, and other tags observed from various models
patterns = [r"(```[^\n]*)\n(.*?)```", r"<([^>]+)>(.*?)"]
@@ -501,6 +536,8 @@ def parseToolCall(answer: str, tool_names: list[str]):
for candidate_dict in candidates:
checked_candidate = checkAndSanitizeToolCallCandidate(candidate_dict, tool_names)
if checked_candidate is not None:
+ if start_pos is None:
+ start_pos = match.start()
matches.append(checked_candidate)
# last resort if nothing has been mapped: LLM might have produced plain json tool call without xml-like tags
@@ -524,4 +561,4 @@ def parseToolCall(answer: str, tool_names: list[str]):
# Ignore invalid JSON silently
pass
- return matches
+ return _return(matches, start_pos)
diff --git a/modules/chat.py b/modules/chat.py
index 8af92273..d3cd0cae 100644
--- a/modules/chat.py
+++ b/modules/chat.py
@@ -23,6 +23,7 @@ from modules.extensions import apply_extensions
from modules.html_generator import (
chat_html_wrapper,
convert_to_markdown,
+ extract_thinking_block,
make_thumbnail
)
from modules.image_utils import open_image_safely
@@ -1168,12 +1169,19 @@ def generate_chat_reply_wrapper(text, state, regenerate=False, _continue=False):
cur_text = text if _tool_turn == 0 else ''
for i, history in enumerate(generate_chat_reply(cur_text, state, regen, cont, loading_message=True, for_ui=True)):
- # Prepend accumulated tool output to visible reply
+ # Prepend accumulated tool output to visible reply for display.
+ # Save and restore the original to prevent the markers from leaking
+ # back into chatbot_wrapper's shared output object, which would cause
+ # duplication on the next yield.
+ _original_visible = history['visible'][-1][1] if visible_prefix else None
if visible_prefix:
- history['visible'][-1][1] = '\n\n'.join(visible_prefix + [history['visible'][-1][1]])
+ history['visible'][-1][1] = '\n\n'.join(visible_prefix + [_original_visible])
yield chat_html_wrapper(history, state['name1'], state['name2'], state['mode'], state['chat_style'], state['character_menu'], last_message_only=(i > 0)), history
+ if visible_prefix:
+ history['visible'][-1][1] = _original_visible
+
if i == 0:
# Save old tool_sequence into version 0 (created by chatbot_wrapper
# on the first yield). Only needed on the first regeneration when
@@ -1196,6 +1204,15 @@ def generate_chat_reply_wrapper(text, state, regenerate=False, _continue=False):
if tool_func_names and parseToolCall(history['internal'][-1][1], tool_func_names):
break
+ # Save the model's visible output before re-applying visible_prefix,
+ # so we can extract thinking content from just this turn's output.
+ _model_visible = history['visible'][-1][1]
+
+ # Re-apply visible prefix to the final state after streaming completes.
+ # This is safe because we're no longer sharing the object with chatbot_wrapper.
+ if visible_prefix:
+ history['visible'][-1][1] = '\n\n'.join(visible_prefix + [_model_visible])
+
save_history(history, state['unique_id'], state['character_menu'], state['mode'])
# Check for tool calls
@@ -1203,7 +1220,7 @@ def generate_chat_reply_wrapper(text, state, regenerate=False, _continue=False):
break
answer = history['internal'][-1][1]
- parsed_calls = parseToolCall(answer, tool_func_names) if answer else None
+ parsed_calls, content_prefix = parseToolCall(answer, tool_func_names, return_prefix=True) if answer else (None, '')
if not parsed_calls:
break # No tool calls — done
@@ -1232,12 +1249,38 @@ def generate_chat_reply_wrapper(text, state, regenerate=False, _continue=False):
# Clear internal (raw tool markup)
history['internal'][-1][1] = ''
- # Add call summary to visible prefix
- call_summary = ', '.join(f'{tc["function"]["name"]}(...)' for tc in parsed_calls)
- visible_prefix.append('Calling: ' + call_summary)
+ # Preserve thinking block and intermediate text from this turn.
+ # content_prefix is the raw text before tool call syntax (returned
+ # by parseToolCall); HTML-escape it and extract thinking to get
+ # the content the user should see.
+ content_text = html.escape(content_prefix)
+ thinking_content, intermediate = extract_thinking_block(content_text)
+ if thinking_content:
+ visible_prefix.append(f'<think>\n{thinking_content}\n</think>')
+ if intermediate and intermediate.strip():
+ visible_prefix.append(intermediate.strip())
- # Execute tools, store results
+ # Build args summaries and show placeholder accordions with "..."
+ # before execution starts (tool calls may be slow, e.g. web search).
+ tc_headers = []
for tc in parsed_calls:
+ fn_name = tc['function']['name']
+ fn_args = tc['function'].get('arguments', {})
+ if isinstance(fn_args, dict) and fn_args:
+ args_summary = ', '.join(f'{k}={json.dumps(v, ensure_ascii=False)}' for k, v in fn_args.items())
+ elif isinstance(fn_args, dict):
+ args_summary = ''
+ else:
+ args_summary = str(fn_args)
+
+ tc_headers.append(f'{fn_name}({args_summary})')
+
+ pending_placeholders = [f'{h}\n...\n' for h in tc_headers]
+ history['visible'][-1][1] = '\n\n'.join(visible_prefix + pending_placeholders)
+ yield chat_html_wrapper(history, state['name1'], state['name2'], state['mode'], state['chat_style'], state['character_menu']), history
+
+ # Execute tools, store results, and replace placeholders with real results
+ for i, tc in enumerate(parsed_calls):
fn_name = tc['function']['name']
fn_args = tc['function'].get('arguments', {})
result = execute_tool(fn_name, fn_args, tool_executors)
@@ -1248,11 +1291,14 @@ def generate_chat_reply_wrapper(text, state, regenerate=False, _continue=False):
except (json.JSONDecodeError, TypeError):
pretty_result = result
- visible_prefix.append(f'**{fn_name}**\n```json\n{pretty_result}\n```')
+ # Replace the placeholder with the real result
+ pending_placeholders[i] = f'{tc_headers[i]}\n{pretty_result}\n'
+ history['visible'][-1][1] = '\n\n'.join(visible_prefix + pending_placeholders)
+ yield chat_html_wrapper(history, state['name1'], state['name2'], state['mode'], state['chat_style'], state['character_menu']), history
- # Show tool results
+ # Move completed tool calls into visible_prefix for next turns
+ visible_prefix.extend(pending_placeholders)
history['visible'][-1][1] = '\n\n'.join(visible_prefix)
- yield chat_html_wrapper(history, state['name1'], state['name2'], state['mode'], state['chat_style'], state['character_menu']), history
save_history(history, state['unique_id'], state['character_menu'], state['mode'])
state['history'] = history
diff --git a/modules/html_generator.py b/modules/html_generator.py
index 4d9904fb..f3811602 100644
--- a/modules/html_generator.py
+++ b/modules/html_generator.py
@@ -114,7 +114,36 @@ def extract_thinking_block(string):
return extract_reasoning(string, html_escaped=True)
-def build_thinking_block(thinking_content, message_id, has_remaining_content):
+
+def build_tool_call_block(header, body, message_id, index):
+ """Build HTML for a tool call accordion block."""
+ block_id = f"tool-call-{message_id}-{index}"
+
+ if body == '...':
+ # Pending placeholder — no expandable body, just title with ellipsis
+ return f'''
+
+
+
+ '''
+
+ # Build a plain directly to avoid highlight.js auto-detection
+ escaped_body = html.escape(body)
+ return f'''
+
+
+
+
+ '''
+
+
+def build_thinking_block(thinking_content, message_id, has_remaining_content, thinking_index=0):
"""Build HTML for a thinking block."""
if thinking_content is None:
return None
@@ -123,7 +152,7 @@ def build_thinking_block(thinking_content, message_id, has_remaining_content):
thinking_html = process_markdown_content(thinking_content)
# Generate unique ID for the thinking block
- block_id = f"thinking-{message_id}-0"
+ block_id = f"thinking-{message_id}-{thinking_index}"
# Check if thinking is complete or still in progress
is_streaming = not has_remaining_content
@@ -304,24 +333,64 @@ def convert_to_markdown(string, message_id=None):
if message_id is None:
message_id = "unknown"
- # Extract different components from the string
- thinking_content, remaining_content = extract_thinking_block(string)
+ # Find tool call blocks by position, then process the text segments
+ # between them using extract_thinking_block (which supports all
+ # THINKING_FORMATS, including end-only variants like Qwen's).
+ tool_call_pattern = re.compile(r'(.*?)\n(.*?)\n', re.DOTALL)
+ tool_calls = list(tool_call_pattern.finditer(string))
- # Build individual HTML blocks
- blocks = []
+ if not tool_calls:
+ # No tool calls — use original single-pass extraction
+ thinking_content, remaining_content = extract_thinking_block(string)
+ blocks = []
+ thinking_html = build_thinking_block(thinking_content, message_id, bool(remaining_content))
+ if thinking_html:
+ blocks.append(thinking_html)
- # Add thinking block if present
- thinking_html = build_thinking_block(thinking_content, message_id, bool(remaining_content))
- if thinking_html:
- blocks.append(thinking_html)
+ main_html = build_main_content_block(remaining_content)
+ if main_html:
+ blocks.append(main_html)
- # Add main content block
- main_html = build_main_content_block(remaining_content)
- if main_html:
- blocks.append(main_html)
+ return ''.join(blocks)
- # Assemble all blocks into final HTML
- return ''.join(blocks)
+ # Split string into text segments around tool_call blocks and
+ # run extract_thinking_block on each segment for full format support.
+ html_parts = []
+ last_end = 0
+ tool_idx = 0
+ think_idx = 0
+
+ def process_text_segment(text, is_last_segment):
+ """Process a text segment between tool_call blocks for thinking content."""
+ nonlocal think_idx
+ if not text.strip():
+ return
+
+ thinking_content, remaining = extract_thinking_block(text)
+ if thinking_content is not None:
+ has_remaining = bool(remaining.strip()) or not is_last_segment
+ html_parts.append(build_thinking_block(thinking_content, message_id, has_remaining, think_idx))
+ think_idx += 1
+ text = remaining
+
+ if text.strip():
+ html_parts.append(process_markdown_content(text))
+
+ for tc in tool_calls:
+ # Process text before this tool_call
+ process_text_segment(string[last_end:tc.start()], is_last_segment=False)
+
+ # Add tool call accordion
+ header = tc.group(1).strip()
+ body = tc.group(2).strip()
+ html_parts.append(build_tool_call_block(header, body, message_id, tool_idx))
+ tool_idx += 1
+ last_end = tc.end()
+
+ # Process text after the last tool_call
+ process_text_segment(string[last_end:], is_last_segment=True)
+
+ return ''.join(html_parts)
def convert_to_markdown_wrapped(string, message_id=None, use_cache=True):
@@ -379,6 +448,7 @@ branch_svg = ''''''
info_svg = ''''''
info_svg_small = ''''''
+tool_svg_small = ''''''
attachment_svg = ''''''
copy_button = f''
From 3e6bd1a310217c89fa5dff23c34e92ae43f1acb9 Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Thu, 12 Mar 2026 14:30:51 -0700
Subject: [PATCH 042/116] UI: Prepend thinking tag when template appends it to
prompt
Makes Qwen models have a thinking block straight away during streaming.
---
modules/chat.py | 14 ++++++++++++++
1 file changed, 14 insertions(+)
diff --git a/modules/chat.py b/modules/chat.py
index d3cd0cae..e4d5dd30 100644
--- a/modules/chat.py
+++ b/modules/chat.py
@@ -28,6 +28,7 @@ from modules.html_generator import (
)
from modules.image_utils import open_image_safely
from modules.logging_colors import logger
+from modules.reasoning import THINKING_FORMATS
from modules.text_generation import (
generate_reply,
get_encoded_length,
@@ -986,10 +987,23 @@ def chatbot_wrapper(text, state, regenerate=False, _continue=False, loading_mess
# Add timestamp for assistant's response at the start of generation
update_message_metadata(output['metadata'], "assistant", row_idx, timestamp=get_current_timestamp(), model_name=shared.model_name)
+ # Detect if the template appended a thinking start tag to the prompt
+ thinking_prefix = None
+ if not _continue:
+ stripped_prompt = prompt.rstrip('\n')
+ for start_tag, end_tag, content_tag in THINKING_FORMATS:
+ if start_tag is not None and stripped_prompt.endswith(start_tag):
+ thinking_prefix = start_tag
+ break
+
# Generate
reply = None
for j, reply in enumerate(generate_reply(prompt, state, stopping_strings=stopping_strings, is_chat=True, for_ui=for_ui)):
+ # Prepend thinking tag if the template appended it to the prompt
+ if thinking_prefix:
+ reply = thinking_prefix + reply
+
# Extract the reply
if state['mode'] in ['chat', 'chat-instruct']:
if not _continue:
From bbd43d9463cded0aae3ce4e1ced1519693bebf9a Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Thu, 12 Mar 2026 14:54:05 -0700
Subject: [PATCH 043/116] UI: Correctly propagate truncation_length when
ctx_size is auto
---
modules/llama_cpp_server.py | 8 +++++++-
modules/models.py | 2 ++
modules/ui_model_menu.py | 6 +++++-
3 files changed, 14 insertions(+), 2 deletions(-)
diff --git a/modules/llama_cpp_server.py b/modules/llama_cpp_server.py
index a3e431ac..192aa9e4 100644
--- a/modules/llama_cpp_server.py
+++ b/modules/llama_cpp_server.py
@@ -36,6 +36,7 @@ class LlamaServer:
self.process = None
self.session = requests.Session()
self.vocabulary_size = None
+ self.n_ctx = None
self.bos_token = ""
self.last_prompt_token_count = 0
@@ -320,12 +321,17 @@ class LlamaServer:
self.vocabulary_size = model_info["meta"]["n_vocab"]
def _get_bos_token(self):
- """Get and store the model's BOS token."""
+ """Get and store the model's BOS token and context size."""
url = f"http://127.0.0.1:{self.port}/props"
response = self.session.get(url).json()
if "bos_token" in response:
self.bos_token = response["bos_token"]
+ # Get actual n_ctx from the server (important when --fit auto-selects it)
+ n_ctx = response.get("default_generation_settings", {}).get("n_ctx")
+ if n_ctx:
+ self.n_ctx = n_ctx
+
def _is_port_available(self, port):
"""Check if a port is available for use."""
with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s:
diff --git a/modules/models.py b/modules/models.py
index 48d68b0b..d83b98d7 100644
--- a/modules/models.py
+++ b/modules/models.py
@@ -54,6 +54,8 @@ def load_model(model_name, loader=None):
if loader.lower().startswith('exllama') or loader.lower().startswith('tensorrt') or loader == 'llama.cpp':
if shared.args.ctx_size > 0:
shared.settings['truncation_length'] = shared.args.ctx_size
+ elif loader == 'llama.cpp' and hasattr(model, 'n_ctx') and model.n_ctx:
+ shared.settings['truncation_length'] = model.n_ctx
shared.is_multimodal = False
if loader.lower() in ('exllamav3', 'llama.cpp') and hasattr(model, 'is_multimodal'):
diff --git a/modules/ui_model_menu.py b/modules/ui_model_menu.py
index 7e91f1ce..5c83096f 100644
--- a/modules/ui_model_menu.py
+++ b/modules/ui_model_menu.py
@@ -388,7 +388,11 @@ def download_model_wrapper(repo_id, specific_file, progress=gr.Progress(), retur
def update_truncation_length(current_length, state):
if 'loader' in state:
if state['loader'].lower().startswith('exllama') or state['loader'] == 'llama.cpp':
- return state['ctx_size']
+ if state['ctx_size'] > 0:
+ return state['ctx_size']
+
+ # ctx_size == 0 means auto: use the actual value from the server
+ return shared.settings['truncation_length']
return current_length
From 4f82b71ef3cfd104cf88c50a7b427063fc2c6094 Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Thu, 12 Mar 2026 14:56:35 -0700
Subject: [PATCH 044/116] UI: Bump the ctx-size max from 131072 to 262144
(256K)
---
modules/ui_model_menu.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/modules/ui_model_menu.py b/modules/ui_model_menu.py
index 5c83096f..d17f586b 100644
--- a/modules/ui_model_menu.py
+++ b/modules/ui_model_menu.py
@@ -42,7 +42,7 @@ def create_ui():
with gr.Row():
with gr.Column():
shared.gradio['gpu_layers'] = gr.Slider(label="gpu-layers", minimum=-1, maximum=get_initial_gpu_layers_max(), step=1, value=shared.args.gpu_layers, info='Number of layers to offload to the GPU. -1 = auto.')
- shared.gradio['ctx_size'] = gr.Slider(label='ctx-size', minimum=0, maximum=131072, step=256, value=shared.args.ctx_size, info='Context length. llama.cpp: 0 = auto if gpu-layers is also -1. Common values: 4096, 8192, 16384, 32768, 65536, 131072.')
+ shared.gradio['ctx_size'] = gr.Slider(label='ctx-size', minimum=0, maximum=262144, step=256, value=shared.args.ctx_size, info='Context length. llama.cpp: 0 = auto if gpu-layers is also -1. Common values: 4096, 8192, 16384, 32768, 65536, 131072.')
shared.gradio['gpu_split'] = gr.Textbox(label='gpu-split', info='Comma-separated list of VRAM (in GB) to use per GPU. Example: 20,7,7')
shared.gradio['attn_implementation'] = gr.Dropdown(label="attn-implementation", choices=['sdpa', 'eager', 'flash_attention_2'], value=shared.args.attn_implementation, info='Attention implementation.')
shared.gradio['cache_type'] = gr.Dropdown(label="cache-type", choices=['fp16', 'q8_0', 'q4_0', 'fp8', 'q8', 'q7', 'q6', 'q5', 'q4', 'q3', 'q2'], value=shared.args.cache_type, allow_custom_value=True, info='Valid options: llama.cpp - fp16, q8_0, q4_0; ExLlamaV3 - fp16, q2 to q8. For ExLlamaV3, you can type custom combinations for separate k/v bits (e.g. q4_q8).')
From fdd8e5b1fd92de8d0ee20e8166b88c8356d13c52 Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Thu, 12 Mar 2026 15:48:50 -0700
Subject: [PATCH 045/116] Make repeated Ctrl+C force a shutdown
---
server.py | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/server.py b/server.py
index 73f190b6..340f7126 100644
--- a/server.py
+++ b/server.py
@@ -33,6 +33,10 @@ warnings.filterwarnings('ignore', category=UserWarning, message='Field "model_na
def signal_handler(sig, frame):
+ # On second Ctrl+C, force an immediate exit
+ signal.signal(signal.SIGINT, signal.SIG_DFL)
+ signal.signal(signal.SIGTERM, signal.SIG_DFL)
+
logger.info("Received Ctrl+C. Shutting down Text Generation Web UI gracefully.")
# Explicitly stop LlamaServer to avoid __del__ cleanup issues during shutdown
From 09d5e049d672e6bddbf6aec402a8b38cd97e074c Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Thu, 12 Mar 2026 16:53:49 -0700
Subject: [PATCH 046/116] UI: Improve the Tools checkbox list style
---
css/main.css | 83 ++++++++++++++++++++++++++++++++++++++++++++++
js/main.js | 17 ++++++++++
modules/ui_chat.py | 5 +--
3 files changed, 103 insertions(+), 2 deletions(-)
diff --git a/css/main.css b/css/main.css
index 2d19b5c8..a7069f33 100644
--- a/css/main.css
+++ b/css/main.css
@@ -1814,3 +1814,86 @@ tr + tr th { border-top: 1px solid; }
thead + tbody tr:first-child td,
thead + tbody tr:first-child th { border-top: 1px solid; }
+
+/* ------------------------------------------------
+ Tools CheckboxGroup - vertical DragDrop-like style
+ ------------------------------------------------ */
+
+/* "Refresh list" link in the Tools label */
+.tools-refresh-link {
+ cursor: pointer;
+}
+
+/* Checkbox list container */
+#tools-group {
+ padding: 0 !important;
+ border-width: 0 !important;
+ background: transparent !important;
+ min-height: 0 !important;
+}
+
+#tools-group .wrap {
+ display: flex;
+ flex-direction: column;
+ flex-wrap: nowrap;
+ gap: 4px;
+ padding: 0;
+ margin-top: var(--spacing-lg);
+ max-height: 350px;
+ overflow-y: auto;
+}
+
+/* Pretty scrollbar for the tools list */
+#tools-group .wrap::-webkit-scrollbar {
+ width: 8px;
+ height: 8px;
+}
+
+#tools-group .wrap::-webkit-scrollbar-track {
+ background: transparent;
+}
+
+#tools-group .wrap::-webkit-scrollbar-thumb,
+#tools-group .wrap::-webkit-scrollbar-thumb:hover {
+ background: var(--neutral-300);
+ border-radius: 30px;
+}
+
+.dark #tools-group .wrap::-webkit-scrollbar-thumb,
+.dark #tools-group .wrap::-webkit-scrollbar-thumb:hover {
+ background: rgb(255 255 255 / 6.25%);
+ border-radius: 10px;
+}
+
+#tools-group .wrap::-webkit-scrollbar-corner {
+ background: transparent;
+}
+
+/* Each checkbox item */
+#tools-group label {
+ display: flex;
+ align-items: center;
+ gap: 8px;
+ padding: 5px 8px;
+ border-radius: var(--radius-sm, 4px);
+ background: var(--block-background-fill);
+ border: 1px solid var(--border-color-primary);
+ color: var(--body-text-color);
+ font-size: var(--input-text-size);
+ font-weight: var(--input-text-weight);
+ cursor: pointer;
+ user-select: none;
+ transition: border-color 0.15s ease, background 0.15s ease;
+ box-shadow: none;
+}
+
+#tools-group label:hover {
+ border-color: var(--input-border-color-focus);
+}
+
+#tools-group label span {
+ flex: 1;
+ overflow: hidden;
+ text-overflow: ellipsis;
+ white-space: nowrap;
+}
diff --git a/js/main.js b/js/main.js
index a7dc77bb..0bc76b2a 100644
--- a/js/main.js
+++ b/js/main.js
@@ -303,6 +303,23 @@ for(i = 0; i < scrollbarElements.length; i++) {
scrollbarElements[i].style.resize = "none";
}
+
+//------------------------------------------------
+// Tools: inject "Refresh list" link into the label
+//------------------------------------------------
+const toolsTitle = document.querySelector("#tools-group > [data-testid='block-info']");
+const toolsInfo = toolsTitle ? toolsTitle.nextElementSibling : null;
+if (toolsInfo) {
+ const refreshLink = document.createElement("span");
+ refreshLink.textContent = " [Refresh list]";
+ refreshLink.className = "tools-refresh-link";
+ refreshLink.addEventListener("click", function(e) {
+ e.preventDefault();
+ document.querySelector("#tools-refresh-btn").click();
+ });
+ toolsInfo.appendChild(refreshLink);
+}
+
//------------------------------------------------
// Remove some backgrounds
//------------------------------------------------
diff --git a/modules/ui_chat.py b/modules/ui_chat.py
index d5b13094..8112d956 100644
--- a/modules/ui_chat.py
+++ b/modules/ui_chat.py
@@ -92,8 +92,9 @@ def create_ui():
gr.HTML("")
from modules.tool_use import get_available_tools
- shared.gradio['selected_tools'] = gr.CheckboxGroup(choices=get_available_tools(), value=[], label='Tools', info='Functions the model can call during generation.')
- ui.create_refresh_button(shared.gradio['selected_tools'], lambda: None, lambda: {'choices': get_available_tools()}, 'refresh-button')
+ shared.gradio['selected_tools'] = gr.CheckboxGroup(choices=get_available_tools(), value=[], label='Tools', info='Functions the model can call during generation.', elem_id='tools-group')
+ shared.gradio['tools_refresh'] = gr.Button('Refresh list', elem_id='tools-refresh-btn', visible=False)
+ shared.gradio['tools_refresh'].click(fn=lambda: gr.update(choices=get_available_tools()), inputs=[], outputs=[shared.gradio['selected_tools']])
gr.HTML("")
From 5c02b7f60369be70b52ccf0432a69380b7f5ab4a Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Thu, 12 Mar 2026 17:08:30 -0700
Subject: [PATCH 047/116] Allow the fetch_webpage tool to return links
---
modules/tool_use.py | 3 ++-
modules/ui_chat.py | 2 +-
modules/web_search.py | 4 ++--
3 files changed, 5 insertions(+), 4 deletions(-)
diff --git a/modules/tool_use.py b/modules/tool_use.py
index cb1e140d..e464d5d4 100644
--- a/modules/tool_use.py
+++ b/modules/tool_use.py
@@ -4,13 +4,14 @@ import random
from modules import shared
from modules.logging_colors import logger
+from modules.utils import natural_keys
def get_available_tools():
"""Return sorted list of tool script names from user_data/tools/*.py."""
tools_dir = shared.user_data_dir / 'tools'
tools_dir.mkdir(parents=True, exist_ok=True)
- return sorted(p.stem for p in tools_dir.glob('*.py'))
+ return sorted((p.stem for p in tools_dir.glob('*.py')), key=natural_keys)
def load_tools(selected_names):
diff --git a/modules/ui_chat.py b/modules/ui_chat.py
index 8112d956..039b9af6 100644
--- a/modules/ui_chat.py
+++ b/modules/ui_chat.py
@@ -92,7 +92,7 @@ def create_ui():
gr.HTML("")
from modules.tool_use import get_available_tools
- shared.gradio['selected_tools'] = gr.CheckboxGroup(choices=get_available_tools(), value=[], label='Tools', info='Functions the model can call during generation.', elem_id='tools-group')
+ shared.gradio['selected_tools'] = gr.CheckboxGroup(choices=get_available_tools(), value=shared.settings.get('selected_tools', []), label='Tools', info='Functions the model can call during generation.', elem_id='tools-group')
shared.gradio['tools_refresh'] = gr.Button('Refresh list', elem_id='tools-refresh-btn', visible=False)
shared.gradio['tools_refresh'].click(fn=lambda: gr.update(choices=get_available_tools()), inputs=[], outputs=[shared.gradio['selected_tools']])
diff --git a/modules/web_search.py b/modules/web_search.py
index 597af4b2..b14cd042 100644
--- a/modules/web_search.py
+++ b/modules/web_search.py
@@ -18,7 +18,7 @@ def get_current_timestamp():
return datetime.now().strftime('%b %d, %Y %H:%M')
-def download_web_page(url, timeout=10):
+def download_web_page(url, timeout=10, include_links=False):
"""
Download a web page and convert its HTML content to structured Markdown text.
"""
@@ -35,7 +35,7 @@ def download_web_page(url, timeout=10):
h = html2text.HTML2Text()
h.body_width = 0
h.ignore_images = True
- h.ignore_links = True
+ h.ignore_links = not include_links
# Convert the HTML to Markdown
markdown_text = h.handle(response.text)
From f8936ec47c95e4147116a2d497e5b94a2d7c99ba Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Thu, 12 Mar 2026 17:10:41 -0700
Subject: [PATCH 048/116] Truncate web_search and fetch_webpage tools to 8192
tokens
---
user_data/tools/web_search.py | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/user_data/tools/web_search.py b/user_data/tools/web_search.py
index 8923eab0..46a45a8c 100644
--- a/user_data/tools/web_search.py
+++ b/user_data/tools/web_search.py
@@ -1,4 +1,4 @@
-from modules.web_search import perform_web_search
+from modules.web_search import perform_web_search, truncate_content_by_tokens
tool = {
"type": "function",
@@ -22,6 +22,6 @@ def execute(arguments):
output = []
for r in results:
if r and r["content"].strip():
- output.append({"title": r["title"], "url": r["url"], "content": r["content"][:4000]})
+ output.append({"title": r["title"], "url": r["url"], "content": truncate_content_by_tokens(r["content"])})
return output if output else [{"error": "No results found."}]
From 1b7e6c57057fdddb57c1585e10c9ad6ccf59e856 Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Thu, 12 Mar 2026 17:11:05 -0700
Subject: [PATCH 049/116] Add the fetch_webpage tool source
---
user_data/tools/fetch_webpage.py | 28 ++++++++++++++++++++++++++++
1 file changed, 28 insertions(+)
create mode 100644 user_data/tools/fetch_webpage.py
diff --git a/user_data/tools/fetch_webpage.py b/user_data/tools/fetch_webpage.py
new file mode 100644
index 00000000..e514491c
--- /dev/null
+++ b/user_data/tools/fetch_webpage.py
@@ -0,0 +1,28 @@
+from modules.web_search import download_web_page, truncate_content_by_tokens
+
+tool = {
+ "type": "function",
+ "function": {
+ "name": "fetch_webpage",
+ "description": "Fetch and read the contents of a web page given its URL. Returns the page content as plain text.",
+ "parameters": {
+ "type": "object",
+ "properties": {
+ "url": {"type": "string", "description": "The URL of the web page to fetch."},
+ },
+ "required": ["url"]
+ }
+ }
+}
+
+
+def execute(arguments):
+ url = arguments.get("url", "")
+ if not url:
+ return {"error": "No URL provided."}
+
+ content = download_web_page(url, include_links=True)
+ if not content or not content.strip():
+ return {"error": f"Failed to fetch content from {url}"}
+
+ return {"url": url, "content": truncate_content_by_tokens(content)}
From a09f21b9de44480d92cb788053852721806c376a Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Thu, 12 Mar 2026 22:17:20 -0300
Subject: [PATCH 050/116] UI: Fix tool calling for GPT-OSS and Continue
---
extensions/openai/utils.py | 7 ++-
modules/chat.py | 88 +++++++++++++++++++++++++-------------
modules/reasoning.py | 1 +
3 files changed, 64 insertions(+), 32 deletions(-)
diff --git a/extensions/openai/utils.py b/extensions/openai/utils.py
index 6fb05f08..eb34ce88 100644
--- a/extensions/openai/utils.py
+++ b/extensions/openai/utils.py
@@ -120,12 +120,14 @@ def _parseChannelToolCalls(answer: str, tool_names: list[str]):
"""Parse channel-based tool calls used by GPT-OSS and similar models.
Format:
+ <|start|>assistant to=functions.func_name<|channel|>commentary json<|message|>{"arg": "value"}
+ or:
<|channel|>commentary to=functions.func_name <|constrain|>json<|message|>{"arg": "value"}
"""
matches = []
start_pos = None
for m in re.finditer(
- r'<\|channel\|>commentary to=functions\.([^<\s]+)\s*(?:<\|constrain\|>json)?<\|message\|>',
+ r'<\|channel\|>\w+ to=functions\.([^<\s]+).*?<\|message\|>',
answer
):
func_name = m.group(1).strip()
@@ -137,7 +139,8 @@ def _parseChannelToolCalls(answer: str, tool_names: list[str]):
try:
arguments = json.loads(json_str)
if start_pos is None:
- start_pos = m.start()
+ prefix = answer.rfind('<|start|>assistant', 0, m.start())
+ start_pos = prefix if prefix != -1 else m.start()
matches.append({
"type": "function",
"function": {
diff --git a/modules/chat.py b/modules/chat.py
index e4d5dd30..19f7d07f 100644
--- a/modules/chat.py
+++ b/modules/chat.py
@@ -81,6 +81,13 @@ jinja_env = ImmutableSandboxedEnvironment(
)
jinja_env.globals["strftime_now"] = strftime_now
+
+def _raise_exception(message):
+ raise ValueError(message)
+
+
+jinja_env.globals["raise_exception"] = _raise_exception
+
_template_cache = {}
@@ -1048,16 +1055,23 @@ def chatbot_wrapper(text, state, regenerate=False, _continue=False, loading_mess
yield output
if _continue:
- # Reprocess the entire internal text for extensions (like translation)
- full_internal = output['internal'][-1][1]
- if state['mode'] in ['chat', 'chat-instruct']:
- full_visible = re.sub("(||{{user}})", state['name1'], full_internal)
- else:
- full_visible = full_internal
+ # Reprocess the entire internal text for extensions (like translation).
+ # Skip the rebuild when the visible text contains markers,
+ # since those only exist in visible (internal is cleared after each tool
+ # execution) and rebuilding from internal would destroy them.
+ if '' not in output['visible'][-1][1]:
+ full_internal = output['internal'][-1][1]
+ if state['mode'] in ['chat', 'chat-instruct']:
+ full_visible = re.sub("(||{{user}})", state['name1'], full_internal)
+ else:
+ full_visible = full_internal
- full_visible = html.escape(full_visible)
- if not state.get('_skip_output_extensions'):
- output['visible'][-1][1] = apply_extensions('output', full_visible, state, is_chat=True)
+ full_visible = html.escape(full_visible)
+ if not state.get('_skip_output_extensions'):
+ output['visible'][-1][1] = apply_extensions('output', full_visible, state, is_chat=True)
+ else:
+ if not state.get('_skip_output_extensions'):
+ output['visible'][-1][1] = apply_extensions('output', output['visible'][-1][1], state, is_chat=True)
else:
if not state.get('_skip_output_extensions'):
output['visible'][-1][1] = apply_extensions('output', output['visible'][-1][1], state, is_chat=True)
@@ -1222,6 +1236,18 @@ def generate_chat_reply_wrapper(text, state, regenerate=False, _continue=False):
# so we can extract thinking content from just this turn's output.
_model_visible = history['visible'][-1][1]
+ # Recover visible_prefix from existing visible text (e.g. on Continue
+ # after a previous session had tool calls). Extract all
+ # blocks and any text between them (thinking blocks, intermediate text).
+ if not visible_prefix and _model_visible:
+ tc_matches = list(re.finditer(r'.*?', _model_visible, re.DOTALL))
+ if tc_matches:
+ prefix_end = tc_matches[-1].end()
+ prefix = _model_visible[:prefix_end].strip()
+ if prefix:
+ visible_prefix = [prefix]
+ _model_visible = _model_visible[prefix_end:].strip()
+
# Re-apply visible prefix to the final state after streaming completes.
# This is safe because we're no longer sharing the object with chatbot_wrapper.
if visible_prefix:
@@ -1244,20 +1270,35 @@ def generate_chat_reply_wrapper(text, state, regenerate=False, _continue=False):
meta = history.get('metadata', {})
seq = meta.setdefault(f'assistant_{row_idx}', {}).setdefault('tool_sequence', [])
- # Serialize tool calls
+ def _render():
+ return chat_html_wrapper(history, state['name1'], state['name2'], state['mode'], state['chat_style'], state['character_menu'])
+
+ # Serialize tool calls and build display headers in one pass
serialized = []
+ tc_headers = []
for tc in parsed_calls:
tc['id'] = generate_tool_call_id()
- args = tc['function'].get('arguments', {})
+ fn_name = tc['function']['name']
+ fn_args = tc['function'].get('arguments', {})
+
serialized.append({
'id': tc['id'],
'type': 'function',
'function': {
- 'name': tc['function']['name'],
- 'arguments': json.dumps(args) if isinstance(args, dict) else args
+ 'name': fn_name,
+ 'arguments': json.dumps(fn_args) if isinstance(fn_args, dict) else fn_args
}
})
+ if isinstance(fn_args, dict) and fn_args:
+ args_summary = ', '.join(f'{k}={json.dumps(v, ensure_ascii=False)}' for k, v in fn_args.items())
+ elif isinstance(fn_args, dict):
+ args_summary = ''
+ else:
+ args_summary = str(fn_args)
+
+ tc_headers.append(f'{fn_name}({args_summary})')
+
seq.append({'tool_calls': serialized})
# Clear internal (raw tool markup)
@@ -1274,24 +1315,11 @@ def generate_chat_reply_wrapper(text, state, regenerate=False, _continue=False):
if intermediate and intermediate.strip():
visible_prefix.append(intermediate.strip())
- # Build args summaries and show placeholder accordions with "..."
- # before execution starts (tool calls may be slow, e.g. web search).
- tc_headers = []
- for tc in parsed_calls:
- fn_name = tc['function']['name']
- fn_args = tc['function'].get('arguments', {})
- if isinstance(fn_args, dict) and fn_args:
- args_summary = ', '.join(f'{k}={json.dumps(v, ensure_ascii=False)}' for k, v in fn_args.items())
- elif isinstance(fn_args, dict):
- args_summary = ''
- else:
- args_summary = str(fn_args)
-
- tc_headers.append(f'{fn_name}({args_summary})')
-
+ # Show placeholder accordions with "..." before execution starts
+ # (tool calls may be slow, e.g. web search).
pending_placeholders = [f'{h}\n...\n' for h in tc_headers]
history['visible'][-1][1] = '\n\n'.join(visible_prefix + pending_placeholders)
- yield chat_html_wrapper(history, state['name1'], state['name2'], state['mode'], state['chat_style'], state['character_menu']), history
+ yield _render(), history
# Execute tools, store results, and replace placeholders with real results
for i, tc in enumerate(parsed_calls):
@@ -1308,7 +1336,7 @@ def generate_chat_reply_wrapper(text, state, regenerate=False, _continue=False):
# Replace the placeholder with the real result
pending_placeholders[i] = f'{tc_headers[i]}\n{pretty_result}\n'
history['visible'][-1][1] = '\n\n'.join(visible_prefix + pending_placeholders)
- yield chat_html_wrapper(history, state['name1'], state['name2'], state['mode'], state['chat_style'], state['character_menu']), history
+ yield _render(), history
# Move completed tool calls into visible_prefix for next turns
visible_prefix.extend(pending_placeholders)
diff --git a/modules/reasoning.py b/modules/reasoning.py
index 12f8553d..708ee55a 100644
--- a/modules/reasoning.py
+++ b/modules/reasoning.py
@@ -5,6 +5,7 @@ import html as html_module
THINKING_FORMATS = [
('', '', None),
('<|channel|>analysis<|message|>', '<|end|>', '<|start|>assistant<|channel|>final<|message|>'),
+ ('<|channel|>commentary<|message|>', '<|end|>', '<|start|>assistant<|channel|>final<|message|>'),
('', '', None),
('<|think|>', '<|end|>', '<|content|>'), # Solar Open
('Thinking Process:', '', None), # Qwen3.5 verbose thinking outside tags
From 4c7a56c18df369d5d04870d6280468900662d3cd Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Thu, 12 Mar 2026 22:17:23 -0300
Subject: [PATCH 051/116] Add num_pages and max_tokens kwargs to web search
tools
---
user_data/tools/fetch_webpage.py | 4 +++-
user_data/tools/web_search.py | 8 ++++++--
2 files changed, 9 insertions(+), 3 deletions(-)
diff --git a/user_data/tools/fetch_webpage.py b/user_data/tools/fetch_webpage.py
index e514491c..ca3e7331 100644
--- a/user_data/tools/fetch_webpage.py
+++ b/user_data/tools/fetch_webpage.py
@@ -9,6 +9,7 @@ tool = {
"type": "object",
"properties": {
"url": {"type": "string", "description": "The URL of the web page to fetch."},
+ "max_tokens": {"type": "integer", "description": "Maximum number of tokens in the returned content (default: 2048)."},
},
"required": ["url"]
}
@@ -18,6 +19,7 @@ tool = {
def execute(arguments):
url = arguments.get("url", "")
+ max_tokens = arguments.get("max_tokens", 2048)
if not url:
return {"error": "No URL provided."}
@@ -25,4 +27,4 @@ def execute(arguments):
if not content or not content.strip():
return {"error": f"Failed to fetch content from {url}"}
- return {"url": url, "content": truncate_content_by_tokens(content)}
+ return {"url": url, "content": truncate_content_by_tokens(content, max_tokens=max_tokens)}
diff --git a/user_data/tools/web_search.py b/user_data/tools/web_search.py
index 46a45a8c..80845963 100644
--- a/user_data/tools/web_search.py
+++ b/user_data/tools/web_search.py
@@ -9,6 +9,8 @@ tool = {
"type": "object",
"properties": {
"query": {"type": "string", "description": "The search query."},
+ "num_pages": {"type": "integer", "description": "Number of search result pages to fetch (default: 3)."},
+ "max_tokens": {"type": "integer", "description": "Maximum number of tokens per page result (default: 2048)."},
},
"required": ["query"]
}
@@ -18,10 +20,12 @@ tool = {
def execute(arguments):
query = arguments.get("query", "")
- results = perform_web_search(query, num_pages=3)
+ num_pages = arguments.get("num_pages", 3)
+ max_tokens = arguments.get("max_tokens", 2048)
+ results = perform_web_search(query, num_pages=num_pages)
output = []
for r in results:
if r and r["content"].strip():
- output.append({"title": r["title"], "url": r["url"], "content": truncate_content_by_tokens(r["content"])})
+ output.append({"title": r["title"], "url": r["url"], "content": truncate_content_by_tokens(r["content"], max_tokens=max_tokens)})
return output if output else [{"error": "No results found."}]
From 286ae475f6794cfb39bf9dabdd289b52eafd5c5b Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Thu, 12 Mar 2026 22:39:38 -0300
Subject: [PATCH 052/116] UI: Clean up tool calling code
---
modules/chat.py | 36 ++++++++++++++++++------------------
modules/tool_use.py | 7 -------
2 files changed, 18 insertions(+), 25 deletions(-)
diff --git a/modules/chat.py b/modules/chat.py
index 19f7d07f..1649a6e8 100644
--- a/modules/chat.py
+++ b/modules/chat.py
@@ -1056,9 +1056,10 @@ def chatbot_wrapper(text, state, regenerate=False, _continue=False, loading_mess
if _continue:
# Reprocess the entire internal text for extensions (like translation).
- # Skip the rebuild when the visible text contains markers,
+ # Skip entirely when the visible text contains markers,
# since those only exist in visible (internal is cleared after each tool
- # execution) and rebuilding from internal would destroy them.
+ # execution) and rebuilding from internal would destroy them. Output
+ # extensions also can't handle the raw markup safely.
if '' not in output['visible'][-1][1]:
full_internal = output['internal'][-1][1]
if state['mode'] in ['chat', 'chat-instruct']:
@@ -1069,9 +1070,6 @@ def chatbot_wrapper(text, state, regenerate=False, _continue=False, loading_mess
full_visible = html.escape(full_visible)
if not state.get('_skip_output_extensions'):
output['visible'][-1][1] = apply_extensions('output', full_visible, state, is_chat=True)
- else:
- if not state.get('_skip_output_extensions'):
- output['visible'][-1][1] = apply_extensions('output', output['visible'][-1][1], state, is_chat=True)
else:
if not state.get('_skip_output_extensions'):
output['visible'][-1][1] = apply_extensions('output', output['visible'][-1][1], state, is_chat=True)
@@ -1141,16 +1139,6 @@ def generate_chat_reply_wrapper(text, state, regenerate=False, _continue=False):
if not character_is_loaded(state):
return
- # On regenerate, clear old tool_sequence metadata so it gets rebuilt.
- # Save it first so it can be stored per-version below.
- _old_tool_sequence = None
- if regenerate:
- history = state['history']
- meta = history.get('metadata', {})
- row_idx = len(history['internal']) - 1
- if row_idx >= 0:
- _old_tool_sequence = meta.get(f'assistant_{row_idx}', {}).pop('tool_sequence', None)
-
if state['start_with'] != '' and not _continue:
if regenerate:
text, state['history'] = remove_last_message(state['history'])
@@ -1160,13 +1148,25 @@ def generate_chat_reply_wrapper(text, state, regenerate=False, _continue=False):
send_dummy_message(text, state)
send_dummy_reply(state['start_with'], state)
+ # On regenerate, clear old tool_sequence metadata so it gets rebuilt.
+ # Save it first so it can be stored per-version below.
+ # This must happen after the start_with logic above, which may remove
+ # and re-add messages, changing which row we operate on.
+ _old_tool_sequence = None
+ if regenerate:
+ history = state['history']
+ meta = history.get('metadata', {})
+ row_idx = len(history['internal']) - 1
+ if row_idx >= 0:
+ _old_tool_sequence = meta.get(f'assistant_{row_idx}', {}).pop('tool_sequence', None)
+
# Load tools if any are selected
selected = state.get('selected_tools', [])
parseToolCall = None
if selected:
- from modules.tool_use import load_tools, execute_tool, generate_tool_call_id
+ from modules.tool_use import load_tools, execute_tool
try:
- from extensions.openai.utils import parseToolCall
+ from extensions.openai.utils import parseToolCall, getToolCallId
except ImportError:
logger.warning('Tool calling requires the openai extension for parseToolCall. Disabling tools.')
selected = []
@@ -1277,7 +1277,7 @@ def generate_chat_reply_wrapper(text, state, regenerate=False, _continue=False):
serialized = []
tc_headers = []
for tc in parsed_calls:
- tc['id'] = generate_tool_call_id()
+ tc['id'] = getToolCallId()
fn_name = tc['function']['name']
fn_args = tc['function'].get('arguments', {})
diff --git a/modules/tool_use.py b/modules/tool_use.py
index e464d5d4..d65c14ed 100644
--- a/modules/tool_use.py
+++ b/modules/tool_use.py
@@ -1,6 +1,5 @@
import importlib.util
import json
-import random
from modules import shared
from modules.logging_colors import logger
@@ -49,12 +48,6 @@ def load_tools(selected_names):
return tool_defs, executors
-def generate_tool_call_id():
- """Generate a unique tool call ID (e.g. 'call_a1b2c3d4')."""
- chars = "abcdefghijklmnopqrstuvwxyz0123456789"
- return "call_" + "".join(random.choice(chars) for _ in range(8))
-
-
def execute_tool(func_name, arguments, executors):
"""Execute a tool by function name. Returns result as a JSON string."""
fn = executors.get(func_name)
From 1ed56aee85b4c64232f3f6517a8b555140f1c8a8 Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Thu, 12 Mar 2026 18:45:19 -0700
Subject: [PATCH 053/116] Add a calculate tool
---
user_data/tools/calculate.py | 48 ++++++++++++++++++++++++++++++++++++
1 file changed, 48 insertions(+)
create mode 100644 user_data/tools/calculate.py
diff --git a/user_data/tools/calculate.py b/user_data/tools/calculate.py
new file mode 100644
index 00000000..e88b71a3
--- /dev/null
+++ b/user_data/tools/calculate.py
@@ -0,0 +1,48 @@
+import ast
+import operator
+
+OPERATORS = {
+ ast.Add: operator.add,
+ ast.Sub: operator.sub,
+ ast.Mult: operator.mul,
+ ast.Div: operator.truediv,
+ ast.Pow: operator.pow,
+ ast.Mod: operator.mod,
+ ast.USub: operator.neg,
+}
+
+
+def _eval(node):
+ if isinstance(node, ast.Constant) and isinstance(node.value, (int, float)):
+ return node.value
+ elif isinstance(node, ast.BinOp) and type(node.op) in OPERATORS:
+ return OPERATORS[type(node.op)](_eval(node.left), _eval(node.right))
+ elif isinstance(node, ast.UnaryOp) and type(node.op) in OPERATORS:
+ return OPERATORS[type(node.op)](_eval(node.operand))
+ raise ValueError(f"Unsupported expression")
+
+
+tool = {
+ "type": "function",
+ "function": {
+ "name": "calculate",
+ "description": "Evaluate a math expression. Supports +, -, *, /, **, %.",
+ "parameters": {
+ "type": "object",
+ "properties": {
+ "expression": {"type": "string", "description": "The math expression to evaluate (e.g. '2 * (3 + 4)')."},
+ },
+ "required": ["expression"]
+ }
+ }
+}
+
+
+def execute(arguments):
+ expr = arguments.get("expression", "")
+ try:
+ tree = ast.parse(expr, mode='eval')
+ result = _eval(tree.body)
+ return {"expression": expr, "result": result}
+ except Exception as e:
+ return {"error": str(e)}
From 0e35421593ebed4295116db35b323ac2e32c91a8 Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Thu, 12 Mar 2026 18:52:41 -0700
Subject: [PATCH 054/116] API: Always extract reasoning_content, even with tool
calls
---
extensions/openai/completions.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/extensions/openai/completions.py b/extensions/openai/completions.py
index 0eb0cd27..290a5bc0 100644
--- a/extensions/openai/completions.py
+++ b/extensions/openai/completions.py
@@ -554,7 +554,7 @@ def chat_completions_common(body: dict, is_legacy: bool = False, stream=False, p
else:
yield chunk
else:
- reasoning, content = extract_reasoning(answer) if not tool_calls else (None, answer)
+ reasoning, content = extract_reasoning(answer)
message = {
"role": "assistant",
"refusal": None,
From 58f26a4cc7d0cec9cc22e4b32bc605540bcdef4f Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Thu, 12 Mar 2026 19:18:55 -0700
Subject: [PATCH 055/116] UI: Skip redundant work in chat loop when no tools
are selected
---
modules/chat.py | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/modules/chat.py b/modules/chat.py
index 1649a6e8..c93972fb 100644
--- a/modules/chat.py
+++ b/modules/chat.py
@@ -1239,7 +1239,7 @@ def generate_chat_reply_wrapper(text, state, regenerate=False, _continue=False):
# Recover visible_prefix from existing visible text (e.g. on Continue
# after a previous session had tool calls). Extract all
# blocks and any text between them (thinking blocks, intermediate text).
- if not visible_prefix and _model_visible:
+ if tool_func_names and not visible_prefix and _model_visible:
tc_matches = list(re.finditer(r'.*?', _model_visible, re.DOTALL))
if tc_matches:
prefix_end = tc_matches[-1].end()
@@ -1253,7 +1253,8 @@ def generate_chat_reply_wrapper(text, state, regenerate=False, _continue=False):
if visible_prefix:
history['visible'][-1][1] = '\n\n'.join(visible_prefix + [_model_visible])
- save_history(history, state['unique_id'], state['character_menu'], state['mode'])
+ if tool_func_names:
+ save_history(history, state['unique_id'], state['character_menu'], state['mode'])
# Check for tool calls
if not tool_func_names or shared.stop_everything:
From 04213dff143cc5a79aaa92fe3ba76271925833ee Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Thu, 12 Mar 2026 19:55:02 -0700
Subject: [PATCH 056/116] Address copilot feedback
---
modules/llama_cpp_server.py | 2 +-
modules/models_settings.py | 3 ++-
modules/text_generation.py | 2 +-
modules/tool_use.py | 3 +++
user_data/tools/calculate.py | 6 +++++-
user_data/tools/roll_dice.py | 4 ++--
6 files changed, 14 insertions(+), 6 deletions(-)
diff --git a/modules/llama_cpp_server.py b/modules/llama_cpp_server.py
index 192aa9e4..1425844d 100644
--- a/modules/llama_cpp_server.py
+++ b/modules/llama_cpp_server.py
@@ -136,7 +136,7 @@ class LlamaServer:
logit_bias = []
if state['custom_token_bans']:
- logit_bias.extend([[int(token_id), False] for token_id in state['custom_token_bans'].split(',')])
+ logit_bias.extend([[int(token_id.strip()), False] for token_id in state['custom_token_bans'].split(',') if token_id.strip()])
if state.get('logit_bias'):
for token_id_str, bias in state['logit_bias'].items():
diff --git a/modules/models_settings.py b/modules/models_settings.py
index 5e69b60e..0e117176 100644
--- a/modules/models_settings.py
+++ b/modules/models_settings.py
@@ -431,7 +431,8 @@ def load_instruction_template(template):
else:
return ''
- file_contents = open(filepath, 'r', encoding='utf-8').read()
+ with open(filepath, 'r', encoding='utf-8') as f:
+ file_contents = f.read()
data = yaml.safe_load(file_contents)
if 'instruction_template' in data:
return data['instruction_template']
diff --git a/modules/text_generation.py b/modules/text_generation.py
index 787c1814..d487cd2f 100644
--- a/modules/text_generation.py
+++ b/modules/text_generation.py
@@ -378,7 +378,7 @@ def generate_reply_HF(question, original_question, state, stopping_strings=None,
generate_params['sampler_priority'] = [x.strip() for x in state['sampler_priority'].replace('\n', ',').split(',') if x.strip()]
if state['custom_token_bans']:
- to_ban = [int(x) for x in state['custom_token_bans'].split(',')]
+ to_ban = [int(x.strip()) for x in state['custom_token_bans'].split(',') if x.strip()]
if len(to_ban) > 0:
if generate_params.get('suppress_tokens', None):
generate_params['suppress_tokens'] += to_ban
diff --git a/modules/tool_use.py b/modules/tool_use.py
index d65c14ed..55424853 100644
--- a/modules/tool_use.py
+++ b/modules/tool_use.py
@@ -42,6 +42,9 @@ def load_tools(selected_names):
continue
func_name = tool_def.get('function', {}).get('name', name)
+ if func_name in executors:
+ logger.warning(f'Tool "{name}" declares function name "{func_name}" which conflicts with an already loaded tool. Skipping.')
+ continue
tool_defs.append(tool_def)
executors[func_name] = execute_fn
diff --git a/user_data/tools/calculate.py b/user_data/tools/calculate.py
index e88b71a3..94f74c41 100644
--- a/user_data/tools/calculate.py
+++ b/user_data/tools/calculate.py
@@ -16,7 +16,11 @@ def _eval(node):
if isinstance(node, ast.Constant) and isinstance(node.value, (int, float)):
return node.value
elif isinstance(node, ast.BinOp) and type(node.op) in OPERATORS:
- return OPERATORS[type(node.op)](_eval(node.left), _eval(node.right))
+ left = _eval(node.left)
+ right = _eval(node.right)
+ if isinstance(node.op, ast.Pow) and isinstance(right, (int, float)) and abs(right) > 10000:
+ raise ValueError("Exponent too large (max 10000)")
+ return OPERATORS[type(node.op)](left, right)
elif isinstance(node, ast.UnaryOp) and type(node.op) in OPERATORS:
return OPERATORS[type(node.op)](_eval(node.operand))
raise ValueError(f"Unsupported expression")
diff --git a/user_data/tools/roll_dice.py b/user_data/tools/roll_dice.py
index 9cab48a8..4af38ddc 100644
--- a/user_data/tools/roll_dice.py
+++ b/user_data/tools/roll_dice.py
@@ -17,7 +17,7 @@ tool = {
def execute(arguments):
- count = arguments.get("count", 1)
- sides = arguments.get("sides", 20)
+ count = max(1, min(arguments.get("count", 1), 1000))
+ sides = max(2, min(arguments.get("sides", 20), 1000))
rolls = [random.randint(1, sides) for _ in range(count)]
return {"rolls": rolls, "total": sum(rolls)}
From 85ec85e569df750724551bf10d5970ee8e7eb351 Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Thu, 12 Mar 2026 20:21:01 -0700
Subject: [PATCH 057/116] UI: Fix Continue while in a tool-calling loop, remove
the upper limit on number of tool calls
---
modules/chat.py | 84 ++++++++++++++++++++++++++++++++++---------------
1 file changed, 59 insertions(+), 25 deletions(-)
diff --git a/modules/chat.py b/modules/chat.py
index c93972fb..b93f2b9d 100644
--- a/modules/chat.py
+++ b/modules/chat.py
@@ -161,6 +161,49 @@ def _deserialize_tool_call_arguments(tool_calls):
return result
+def _expand_tool_sequence(tool_seq):
+ """Expand a tool_sequence list into API messages.
+
+ Returns a list of dicts (role: assistant with tool_calls, or role: tool).
+ If any tool_call IDs are missing a matching tool result, a synthetic
+ empty result is inserted so the prompt is never malformed.
+ """
+ messages = []
+ expected_ids = []
+ seen_ids = set()
+
+ for item in tool_seq:
+ if 'tool_calls' in item:
+ deserialized = _deserialize_tool_call_arguments(item['tool_calls'])
+ messages.append({
+ "role": "assistant",
+ "content": "",
+ "tool_calls": deserialized
+ })
+ for tc in item['tool_calls']:
+ tc_id = tc.get('id', '')
+ if tc_id:
+ expected_ids.append(tc_id)
+ elif item.get('role') == 'tool':
+ messages.append({
+ "role": "tool",
+ "content": item['content'],
+ "tool_call_id": item.get('tool_call_id', '')
+ })
+ seen_ids.add(item.get('tool_call_id', ''))
+
+ # Fill in synthetic results for any orphaned tool call IDs
+ for tc_id in expected_ids:
+ if tc_id not in seen_ids:
+ messages.append({
+ "role": "tool",
+ "content": "",
+ "tool_call_id": tc_id
+ })
+
+ return messages
+
+
def generate_chat_prompt(user_input, state, **kwargs):
impersonate = kwargs.get('impersonate', False)
_continue = kwargs.get('_continue', False)
@@ -312,17 +355,8 @@ def generate_chat_prompt(user_input, state, **kwargs):
meta_key = f"assistant_{row_idx}"
tool_seq = metadata.get(meta_key, {}).get('tool_sequence', [])
if tool_seq:
- for item in reversed(tool_seq):
- if 'tool_calls' in item:
- messages.insert(insert_pos, {
- "role": "assistant", "content": "",
- "tool_calls": _deserialize_tool_call_arguments(item['tool_calls'])
- })
- elif item.get('role') == 'tool':
- messages.insert(insert_pos, {
- "role": "tool", "content": item['content'],
- "tool_call_id": item.get('tool_call_id', '')
- })
+ for msg in reversed(_expand_tool_sequence(tool_seq)):
+ messages.insert(insert_pos, msg)
if entry_meta.get('role') == 'system':
if user_msg:
@@ -400,17 +434,7 @@ def generate_chat_prompt(user_input, state, **kwargs):
# history loop during regenerate — needed so the model sees prior
# tool calls and results when re-generating the final answer).
current_tool_seq = metadata.get(f"assistant_{len(history)}", {}).get('tool_sequence', [])
- for item in current_tool_seq:
- if 'tool_calls' in item:
- messages.append({
- "role": "assistant", "content": "",
- "tool_calls": _deserialize_tool_call_arguments(item['tool_calls'])
- })
- elif item.get('role') == 'tool':
- messages.append({
- "role": "tool", "content": item['content'],
- "tool_call_id": item.get('tool_call_id', '')
- })
+ messages.extend(_expand_tool_sequence(current_tool_seq))
if impersonate and state['mode'] != 'chat-instruct':
messages.append({"role": "user", "content": "fake user message replace me"})
@@ -1181,9 +1205,8 @@ def generate_chat_reply_wrapper(text, state, regenerate=False, _continue=False):
visible_prefix = [] # Accumulated tool call summaries + results
last_save_time = time.monotonic()
save_interval = 8
- max_tool_turns = 10
-
- for _tool_turn in range(max_tool_turns):
+ _tool_turn = 0
+ while True:
history = state['history']
# Turn 0: use original flags; turns 2+: regenerate into the same entry.
@@ -1324,6 +1347,16 @@ def generate_chat_reply_wrapper(text, state, regenerate=False, _continue=False):
# Execute tools, store results, and replace placeholders with real results
for i, tc in enumerate(parsed_calls):
+ # Check for stop request before each tool execution
+ if shared.stop_everything:
+ for j in range(i, len(parsed_calls)):
+ seq.append({'role': 'tool', 'content': 'Tool execution was cancelled by the user.', 'tool_call_id': parsed_calls[j]['id']})
+ pending_placeholders[j] = f'{tc_headers[j]}\nCancelled\n'
+
+ history['visible'][-1][1] = '\n\n'.join(visible_prefix + pending_placeholders)
+ yield _render(), history
+ break
+
fn_name = tc['function']['name']
fn_args = tc['function'].get('arguments', {})
result = execute_tool(fn_name, fn_args, tool_executors)
@@ -1345,6 +1378,7 @@ def generate_chat_reply_wrapper(text, state, regenerate=False, _continue=False):
save_history(history, state['unique_id'], state['character_menu'], state['mode'])
state['history'] = history
+ _tool_turn += 1
state.pop('_tool_turn', None)
state['history'] = history
From c094bc943c5912d592b0f796b75989af4fc82dc4 Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Thu, 12 Mar 2026 21:45:38 -0700
Subject: [PATCH 058/116] UI: Skip output extensions on intermediate
tool-calling turns
---
modules/chat.py | 13 +++++++++++++
1 file changed, 13 insertions(+)
diff --git a/modules/chat.py b/modules/chat.py
index b93f2b9d..57fd50e0 100644
--- a/modules/chat.py
+++ b/modules/chat.py
@@ -1214,6 +1214,7 @@ def generate_chat_reply_wrapper(text, state, regenerate=False, _continue=False):
# that intermediate tool-loop regenerations don't pollute swipe history.
if _tool_turn > 0:
state['_tool_turn'] = True
+ state['_skip_output_extensions'] = True
regen = regenerate if _tool_turn == 0 else True
cont = _continue if _tool_turn == 0 else False
@@ -1381,6 +1382,18 @@ def generate_chat_reply_wrapper(text, state, regenerate=False, _continue=False):
_tool_turn += 1
state.pop('_tool_turn', None)
+
+ # If output extensions were deferred during tool turns, apply them now
+ # to the final model response only (not to tool call markers).
+ if state.pop('_skip_output_extensions', None):
+ _model_visible = apply_extensions('output', _model_visible, state, is_chat=True)
+ if visible_prefix:
+ history['visible'][-1][1] = '\n\n'.join(visible_prefix + [_model_visible])
+ else:
+ history['visible'][-1][1] = _model_visible
+
+ yield chat_html_wrapper(history, state['name1'], state['name2'], state['mode'], state['chat_style'], state['character_menu']), history
+
state['history'] = history
# Sync version metadata so swipes show the full visible (with tool prefix)
From 5ddc1002d2f3d6cc8f65aad959dc810e92530319 Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Fri, 13 Mar 2026 02:40:09 -0700
Subject: [PATCH 059/116] Update ExLlamaV3 to 0.0.25
---
requirements/full/requirements.txt | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/requirements/full/requirements.txt b/requirements/full/requirements.txt
index 12e7fbae..8e095fae 100644
--- a/requirements/full/requirements.txt
+++ b/requirements/full/requirements.txt
@@ -42,7 +42,7 @@ tiktoken
# CUDA wheels
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+cu124-py3-none-win_amd64.whl; platform_system == "Windows"
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+cu124-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
-https://github.com/turboderp-org/exllamav3/releases/download/v0.0.24/exllamav3-0.0.24+cu128.torch2.9.0-cp313-cp313-win_amd64.whl; platform_system == "Windows" and python_version == "3.13"
-https://github.com/turboderp-org/exllamav3/releases/download/v0.0.24/exllamav3-0.0.24+cu128.torch2.9.0-cp313-cp313-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.13"
+https://github.com/turboderp-org/exllamav3/releases/download/v0.0.25/exllamav3-0.0.25+cu128.torch2.9.0-cp313-cp313-win_amd64.whl; platform_system == "Windows" and python_version == "3.13"
+https://github.com/turboderp-org/exllamav3/releases/download/v0.0.25/exllamav3-0.0.25+cu128.torch2.9.0-cp313-cp313-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.13"
https://github.com/kingbri1/flash-attention/releases/download/v2.8.3/flash_attn-2.8.3+cu128torch2.9.0cxx11abiFALSE-cp313-cp313-win_amd64.whl; platform_system == "Windows" and python_version == "3.13"
https://github.com/kingbri1/flash-attention/releases/download/v2.8.3/flash_attn-2.8.3+cu128torch2.9.0cxx11abiFALSE-cp313-cp313-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.13"
From a4bef860b6d743c78d0aed36c934cc022df37bee Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Fri, 13 Mar 2026 06:45:29 -0300
Subject: [PATCH 060/116] UI: Optimize chat streaming by batching morphdom to
one update per animation frame
The monitor physically cannot paint faster than its refresh rate, so
intermediate morphdom calls between frames do redundant parsing, diffing,
and patching work that is never displayed.
---
js/global_scope_js.js | 14 ++++++++++++++
1 file changed, 14 insertions(+)
diff --git a/js/global_scope_js.js b/js/global_scope_js.js
index 62b31d37..3207a681 100644
--- a/js/global_scope_js.js
+++ b/js/global_scope_js.js
@@ -269,7 +269,21 @@ function removeLastClick() {
document.getElementById("Remove-last").click();
}
+let pendingMorphdomData = null;
+let morphdomRafId = null;
+
function handleMorphdomUpdate(data) {
+ pendingMorphdomData = data;
+ if (!morphdomRafId) {
+ morphdomRafId = requestAnimationFrame(() => {
+ morphdomRafId = null;
+ applyMorphdomUpdate(pendingMorphdomData);
+ pendingMorphdomData = null;
+ });
+ }
+}
+
+function applyMorphdomUpdate(data) {
// Determine target element and use it as query scope
var target_element, target_html;
if (data.last_message_only) {
From 5833d94d7fad6739ef72d3c169143fbc977211c2 Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Fri, 13 Mar 2026 02:56:49 -0700
Subject: [PATCH 061/116] UI: Prevent word breaks in tables
---
css/main.css | 10 ++++++++++
1 file changed, 10 insertions(+)
diff --git a/css/main.css b/css/main.css
index a7069f33..30c0d5f5 100644
--- a/css/main.css
+++ b/css/main.css
@@ -1802,6 +1802,16 @@ table {
border-collapse: collapse;
}
+.message-body table {
+ display: block;
+ overflow-x: auto;
+}
+
+.message-body :is(td, th) {
+ word-break: normal;
+ overflow-wrap: normal;
+}
+
table, tr, td, th, thead {
border: 0;
}
From fef95b9e56855707cc13e5dc1220193352bc4bc1 Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Fri, 13 Mar 2026 03:05:09 -0700
Subject: [PATCH 062/116] UI: Fix an autoscroll race condition during chat
streaming
---
js/main.js | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/js/main.js b/js/main.js
index 0bc76b2a..135aa948 100644
--- a/js/main.js
+++ b/js/main.js
@@ -145,6 +145,7 @@ targetElement.classList.add("pretty_scrollbar");
targetElement.classList.add("chat-parent");
window.isScrolled = false;
let scrollTimeout;
+let isProgrammaticScroll = false;
targetElement.addEventListener("scroll", function() {
let diff = targetElement.scrollHeight - targetElement.clientHeight;
@@ -157,9 +158,10 @@ targetElement.addEventListener("scroll", function() {
if(isAtBottomNow) {
window.isScrolled = false;
- } else {
+ } else if (!isProgrammaticScroll) {
window.isScrolled = true;
}
+ isProgrammaticScroll = false;
// Clear previous timeout and set new one
clearTimeout(scrollTimeout);
@@ -193,6 +195,7 @@ const observer = new MutationObserver(function(mutations) {
if (!window.isScrolled && !isScrollingClassOnly) {
const maxScroll = targetElement.scrollHeight - targetElement.clientHeight;
if (maxScroll > 0 && targetElement.scrollTop < maxScroll - 1) {
+ isProgrammaticScroll = true;
targetElement.scrollTop = maxScroll;
}
}
@@ -1091,6 +1094,7 @@ document.fonts.addEventListener("loadingdone", (event) => {
if (!window.isScrolled) {
const maxScroll = targetElement.scrollHeight - targetElement.clientHeight;
if (targetElement.scrollTop < maxScroll - 5) {
+ isProgrammaticScroll = true;
targetElement.scrollTop = maxScroll;
}
}
From 46288256511c0e42fb9fb4cfeb8585a2a9362947 Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Fri, 13 Mar 2026 03:17:36 -0700
Subject: [PATCH 063/116] Better solution to
fef95b9e56855707cc13e5dc1220193352bc4bc1
---
js/main.js | 10 +++++-----
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/js/main.js b/js/main.js
index 135aa948..c3e51c3c 100644
--- a/js/main.js
+++ b/js/main.js
@@ -145,7 +145,8 @@ targetElement.classList.add("pretty_scrollbar");
targetElement.classList.add("chat-parent");
window.isScrolled = false;
let scrollTimeout;
-let isProgrammaticScroll = false;
+let lastScrollTop = 0;
+let lastScrollHeight = 0;
targetElement.addEventListener("scroll", function() {
let diff = targetElement.scrollHeight - targetElement.clientHeight;
@@ -158,10 +159,11 @@ targetElement.addEventListener("scroll", function() {
if(isAtBottomNow) {
window.isScrolled = false;
- } else if (!isProgrammaticScroll) {
+ } else if (targetElement.scrollTop < lastScrollTop && targetElement.scrollHeight >= lastScrollHeight) {
window.isScrolled = true;
}
- isProgrammaticScroll = false;
+ lastScrollTop = targetElement.scrollTop;
+ lastScrollHeight = targetElement.scrollHeight;
// Clear previous timeout and set new one
clearTimeout(scrollTimeout);
@@ -195,7 +197,6 @@ const observer = new MutationObserver(function(mutations) {
if (!window.isScrolled && !isScrollingClassOnly) {
const maxScroll = targetElement.scrollHeight - targetElement.clientHeight;
if (maxScroll > 0 && targetElement.scrollTop < maxScroll - 1) {
- isProgrammaticScroll = true;
targetElement.scrollTop = maxScroll;
}
}
@@ -1094,7 +1095,6 @@ document.fonts.addEventListener("loadingdone", (event) => {
if (!window.isScrolled) {
const maxScroll = targetElement.scrollHeight - targetElement.clientHeight;
if (targetElement.scrollTop < maxScroll - 5) {
- isProgrammaticScroll = true;
targetElement.scrollTop = maxScroll;
}
}
From c39c187f47b79b540b8b0abf18979b5103f48bd0 Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Fri, 13 Mar 2026 03:21:47 -0700
Subject: [PATCH 064/116] UI: Improve the style of table scrollbars
---
css/main.css | 3 +--
modules/html_generator.py | 3 +++
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/css/main.css b/css/main.css
index 30c0d5f5..7cc496a7 100644
--- a/css/main.css
+++ b/css/main.css
@@ -1802,8 +1802,7 @@ table {
border-collapse: collapse;
}
-.message-body table {
- display: block;
+.table-wrapper {
overflow-x: auto;
}
diff --git a/modules/html_generator.py b/modules/html_generator.py
index f3811602..138d4ade 100644
--- a/modules/html_generator.py
+++ b/modules/html_generator.py
@@ -317,6 +317,9 @@ def process_markdown_content(string):
# Unescape backslashes
html_output = html_output.replace('\\\\', '\\')
+ # Wrap tables in a scrollable div
+ html_output = html_output.replace('')
+
return html_output
From d0b72c73c08cc74c7d05c5afdbbef7f96b94dcfd Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Fri, 13 Mar 2026 03:43:02 -0700
Subject: [PATCH 065/116] Update diffusers to 0.37
---
requirements/full/requirements.txt | 2 +-
requirements/full/requirements_amd.txt | 2 +-
requirements/full/requirements_apple_intel.txt | 2 +-
requirements/full/requirements_apple_silicon.txt | 2 +-
requirements/full/requirements_cpu_only.txt | 2 +-
requirements/full/requirements_nowheels.txt | 2 +-
6 files changed, 6 insertions(+), 6 deletions(-)
diff --git a/requirements/full/requirements.txt b/requirements/full/requirements.txt
index 8e095fae..03f4abac 100644
--- a/requirements/full/requirements.txt
+++ b/requirements/full/requirements.txt
@@ -2,7 +2,7 @@ accelerate==1.12.*
audioop-lts<1.0; python_version >= "3.13"
bitsandbytes==0.49.*
datasets
-diffusers==0.36.*
+diffusers==0.37.*
einops
fastapi==0.112.4
flash-linear-attention==0.4.*
diff --git a/requirements/full/requirements_amd.txt b/requirements/full/requirements_amd.txt
index 19cc0d9d..f3551fa2 100644
--- a/requirements/full/requirements_amd.txt
+++ b/requirements/full/requirements_amd.txt
@@ -1,7 +1,7 @@
accelerate==1.12.*
audioop-lts<1.0; python_version >= "3.13"
datasets
-diffusers==0.36.*
+diffusers==0.37.*
einops
fastapi==0.112.4
html2text==2025.4.15
diff --git a/requirements/full/requirements_apple_intel.txt b/requirements/full/requirements_apple_intel.txt
index ebe26f9d..5e0cf8ad 100644
--- a/requirements/full/requirements_apple_intel.txt
+++ b/requirements/full/requirements_apple_intel.txt
@@ -1,7 +1,7 @@
accelerate==1.12.*
audioop-lts<1.0; python_version >= "3.13"
datasets
-diffusers==0.36.*
+diffusers==0.37.*
einops
fastapi==0.112.4
html2text==2025.4.15
diff --git a/requirements/full/requirements_apple_silicon.txt b/requirements/full/requirements_apple_silicon.txt
index 49155690..d55c3e24 100644
--- a/requirements/full/requirements_apple_silicon.txt
+++ b/requirements/full/requirements_apple_silicon.txt
@@ -1,7 +1,7 @@
accelerate==1.12.*
audioop-lts<1.0; python_version >= "3.13"
datasets
-diffusers==0.36.*
+diffusers==0.37.*
einops
fastapi==0.112.4
html2text==2025.4.15
diff --git a/requirements/full/requirements_cpu_only.txt b/requirements/full/requirements_cpu_only.txt
index 1c7c5735..34e864ac 100644
--- a/requirements/full/requirements_cpu_only.txt
+++ b/requirements/full/requirements_cpu_only.txt
@@ -1,7 +1,7 @@
accelerate==1.12.*
audioop-lts<1.0; python_version >= "3.13"
datasets
-diffusers==0.36.*
+diffusers==0.37.*
einops
fastapi==0.112.4
html2text==2025.4.15
diff --git a/requirements/full/requirements_nowheels.txt b/requirements/full/requirements_nowheels.txt
index 63823db8..6128c0ed 100644
--- a/requirements/full/requirements_nowheels.txt
+++ b/requirements/full/requirements_nowheels.txt
@@ -1,7 +1,7 @@
accelerate==1.12.*
audioop-lts<1.0; python_version >= "3.13"
datasets
-diffusers==0.36.*
+diffusers==0.37.*
einops
fastapi==0.112.4
html2text==2025.4.15
From b7670cc762206445e4e0ffed76f535c661bb85f4 Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Fri, 13 Mar 2026 04:00:30 -0700
Subject: [PATCH 066/116] Add a tool calling tutorial
---
docs/Tool Calling Tutorial.md | 144 ++++++++++++++++++++++++++++++++++
1 file changed, 144 insertions(+)
create mode 100644 docs/Tool Calling Tutorial.md
diff --git a/docs/Tool Calling Tutorial.md b/docs/Tool Calling Tutorial.md
new file mode 100644
index 00000000..170bdff7
--- /dev/null
+++ b/docs/Tool Calling Tutorial.md
@@ -0,0 +1,144 @@
+## Tool calling in the UI
+
+### 1. Load a model with tool-calling support
+
+Load a model with tool-calling support (Qwen, Mistral, Llama 4, etc.) from the Model tab.
+
+### 2. Select tools
+
+In the chat sidebar, check the tools you want the model to use:
+
+- **web_search** -- Search the web using DuckDuckGo.
+- **fetch_webpage** -- Fetch the content of a URL.
+- **calculate** -- Evaluate math expressions.
+- **get_datetime** -- Get the current date and time.
+- **roll_dice** -- Roll dice.
+
+### 3. Chat
+
+Send a message as usual. When the model decides it needs a tool, it will call it automatically. You will see each tool call and its result in a collapsible accordion inside the chat message.
+
+The model may call multiple tools in sequence before giving its final answer.
+
+## Writing custom tools
+
+Each tool is a single `.py` file in `user_data/tools/`. It needs two things:
+
+1. A `tool` dictionary that describes the function (name, description, parameters).
+2. An `execute(arguments)` function that runs it and returns the result.
+
+Here is a minimal example (`user_data/tools/get_datetime.py`):
+
+```python
+from datetime import datetime
+
+tool = {
+ "type": "function",
+ "function": {
+ "name": "get_datetime",
+ "description": "Get the current date and time.",
+ "parameters": {
+ "type": "object",
+ "properties": {},
+ }
+ }
+}
+
+
+def execute(arguments):
+ now = datetime.now()
+ return {"date": now.strftime("%Y-%m-%d"), "time": now.strftime("%I:%M %p")}
+```
+
+An example with parameters (`user_data/tools/roll_dice.py`):
+
+```python
+import random
+
+tool = {
+ "type": "function",
+ "function": {
+ "name": "roll_dice",
+ "description": "Roll one or more dice with the specified number of sides.",
+ "parameters": {
+ "type": "object",
+ "properties": {
+ "count": {"type": "integer", "description": "Number of dice to roll.", "default": 1},
+ "sides": {"type": "integer", "description": "Number of sides per die.", "default": 20},
+ },
+ }
+ }
+}
+
+
+def execute(arguments):
+ count = max(1, min(arguments.get("count", 1), 1000))
+ sides = max(2, min(arguments.get("sides", 20), 1000))
+ rolls = [random.randint(1, sides) for _ in range(count)]
+ return {"rolls": rolls, "total": sum(rolls)}
+```
+
+You can open the built-in tools in `user_data/tools/` for more examples.
+
+## Tool calling over the API
+
+Tool calling over the API follows the [OpenAI API](https://platform.openai.com/docs/guides/function-calling) convention. Define your tools, send them with your messages, and handle tool calls in a loop until the model gives a final answer.
+
+```python
+import json
+import requests
+
+url = "http://127.0.0.1:5000/v1/chat/completions"
+
+tools = [
+ {
+ "type": "function",
+ "function": {
+ "name": "get_weather",
+ "description": "Get the current weather for a given location.",
+ "parameters": {
+ "type": "object",
+ "properties": {
+ "location": {"type": "string", "description": "City name"},
+ },
+ "required": ["location"]
+ }
+ }
+ }
+]
+
+
+def execute_tool(name, arguments):
+ if name == "get_weather":
+ return {"temperature": "14°C", "condition": "partly cloudy"}
+ return {"error": f"Unknown tool: {name}"}
+
+
+messages = [{"role": "user", "content": "What's the weather like in Paris?"}]
+
+for _ in range(10):
+ response = requests.post(url, json={"messages": messages, "tools": tools}).json()
+ choice = response["choices"][0]
+
+ if choice["finish_reason"] == "tool_calls":
+ messages.append({
+ "role": "assistant",
+ "content": choice["message"]["content"],
+ "tool_calls": choice["message"]["tool_calls"],
+ })
+
+ for tool_call in choice["message"]["tool_calls"]:
+ name = tool_call["function"]["name"]
+ arguments = json.loads(tool_call["function"]["arguments"])
+ result = execute_tool(name, arguments)
+ print(f"Tool call: {name}({arguments}) => {result}")
+
+ messages.append({
+ "role": "tool",
+ "tool_call_id": tool_call["id"],
+ "content": json.dumps(result),
+ })
+ else:
+ print(f"\nAssistant: {choice['message']['content']}")
+ break
+```
From e50b823eee369f50981fcaaef3bf05c8dc3e350d Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Fri, 13 Mar 2026 06:22:28 -0700
Subject: [PATCH 067/116] Update llama.cpp
---
requirements/full/requirements.txt | 4 ++--
requirements/full/requirements_amd.txt | 4 ++--
requirements/full/requirements_apple_intel.txt | 2 +-
requirements/full/requirements_apple_silicon.txt | 2 +-
requirements/full/requirements_cpu_only.txt | 4 ++--
requirements/portable/requirements.txt | 4 ++--
requirements/portable/requirements_amd.txt | 4 ++--
requirements/portable/requirements_apple_intel.txt | 2 +-
requirements/portable/requirements_apple_silicon.txt | 2 +-
requirements/portable/requirements_cpu_only.txt | 4 ++--
requirements/portable/requirements_cuda131.txt | 4 ++--
requirements/portable/requirements_vulkan.txt | 4 ++--
12 files changed, 20 insertions(+), 20 deletions(-)
diff --git a/requirements/full/requirements.txt b/requirements/full/requirements.txt
index 03f4abac..c702a8d3 100644
--- a/requirements/full/requirements.txt
+++ b/requirements/full/requirements.txt
@@ -40,8 +40,8 @@ sse-starlette==1.6.5
tiktoken
# CUDA wheels
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+cu124-py3-none-win_amd64.whl; platform_system == "Windows"
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+cu124-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+cu124-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+cu124-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
https://github.com/turboderp-org/exllamav3/releases/download/v0.0.25/exllamav3-0.0.25+cu128.torch2.9.0-cp313-cp313-win_amd64.whl; platform_system == "Windows" and python_version == "3.13"
https://github.com/turboderp-org/exllamav3/releases/download/v0.0.25/exllamav3-0.0.25+cu128.torch2.9.0-cp313-cp313-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.13"
https://github.com/kingbri1/flash-attention/releases/download/v2.8.3/flash_attn-2.8.3+cu128torch2.9.0cxx11abiFALSE-cp313-cp313-win_amd64.whl; platform_system == "Windows" and python_version == "3.13"
diff --git a/requirements/full/requirements_amd.txt b/requirements/full/requirements_amd.txt
index f3551fa2..65a9aa00 100644
--- a/requirements/full/requirements_amd.txt
+++ b/requirements/full/requirements_amd.txt
@@ -37,5 +37,5 @@ sse-starlette==1.6.5
tiktoken
# AMD wheels
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+rocm7.2-py3-none-win_amd64.whl; platform_system == "Windows"
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+rocm7.2-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+rocm7.2-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+rocm7.2-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
diff --git a/requirements/full/requirements_apple_intel.txt b/requirements/full/requirements_apple_intel.txt
index 5e0cf8ad..bba62491 100644
--- a/requirements/full/requirements_apple_intel.txt
+++ b/requirements/full/requirements_apple_intel.txt
@@ -37,4 +37,4 @@ sse-starlette==1.6.5
tiktoken
# Mac wheels
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0-py3-none-macosx_13_0_x86_64.whl; platform_system == "Darwin"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0-py3-none-macosx_13_0_x86_64.whl; platform_system == "Darwin"
diff --git a/requirements/full/requirements_apple_silicon.txt b/requirements/full/requirements_apple_silicon.txt
index d55c3e24..61dbf51b 100644
--- a/requirements/full/requirements_apple_silicon.txt
+++ b/requirements/full/requirements_apple_silicon.txt
@@ -37,4 +37,4 @@ sse-starlette==1.6.5
tiktoken
# Mac wheels
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0-py3-none-macosx_13_0_arm64.whl; platform_system == "Darwin"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0-py3-none-macosx_13_0_arm64.whl; platform_system == "Darwin"
diff --git a/requirements/full/requirements_cpu_only.txt b/requirements/full/requirements_cpu_only.txt
index 34e864ac..384a552a 100644
--- a/requirements/full/requirements_cpu_only.txt
+++ b/requirements/full/requirements_cpu_only.txt
@@ -37,5 +37,5 @@ sse-starlette==1.6.5
tiktoken
# llama.cpp (CPU only)
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+cpu-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+cpu-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+cpu-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+cpu-py3-none-win_amd64.whl; platform_system == "Windows"
diff --git a/requirements/portable/requirements.txt b/requirements/portable/requirements.txt
index db23d4bf..0e3d67d3 100644
--- a/requirements/portable/requirements.txt
+++ b/requirements/portable/requirements.txt
@@ -23,5 +23,5 @@ sse-starlette==1.6.5
tiktoken
# CUDA wheels
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+cu124-py3-none-win_amd64.whl; platform_system == "Windows"
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+cu124-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+cu124-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+cu124-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
diff --git a/requirements/portable/requirements_amd.txt b/requirements/portable/requirements_amd.txt
index e8cd9fd9..729829b3 100644
--- a/requirements/portable/requirements_amd.txt
+++ b/requirements/portable/requirements_amd.txt
@@ -23,5 +23,5 @@ sse-starlette==1.6.5
tiktoken
# AMD wheels
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+rocm7.2-py3-none-win_amd64.whl; platform_system == "Windows"
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+rocm7.2-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+rocm7.2-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+rocm7.2-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
diff --git a/requirements/portable/requirements_apple_intel.txt b/requirements/portable/requirements_apple_intel.txt
index 24c558a9..4b16414c 100644
--- a/requirements/portable/requirements_apple_intel.txt
+++ b/requirements/portable/requirements_apple_intel.txt
@@ -23,4 +23,4 @@ sse-starlette==1.6.5
tiktoken
# Mac wheels
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0-py3-none-macosx_13_0_x86_64.whl; platform_system == "Darwin"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0-py3-none-macosx_13_0_x86_64.whl; platform_system == "Darwin"
diff --git a/requirements/portable/requirements_apple_silicon.txt b/requirements/portable/requirements_apple_silicon.txt
index f2e8e691..3a1764dc 100644
--- a/requirements/portable/requirements_apple_silicon.txt
+++ b/requirements/portable/requirements_apple_silicon.txt
@@ -23,4 +23,4 @@ sse-starlette==1.6.5
tiktoken
# Mac wheels
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0-py3-none-macosx_13_0_arm64.whl; platform_system == "Darwin"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0-py3-none-macosx_13_0_arm64.whl; platform_system == "Darwin"
diff --git a/requirements/portable/requirements_cpu_only.txt b/requirements/portable/requirements_cpu_only.txt
index 296c0432..9d115c86 100644
--- a/requirements/portable/requirements_cpu_only.txt
+++ b/requirements/portable/requirements_cpu_only.txt
@@ -23,5 +23,5 @@ sse-starlette==1.6.5
tiktoken
# llama.cpp (CPU only)
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+cpu-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+cpu-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+cpu-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+cpu-py3-none-win_amd64.whl; platform_system == "Windows"
diff --git a/requirements/portable/requirements_cuda131.txt b/requirements/portable/requirements_cuda131.txt
index aefce769..4472e1d4 100644
--- a/requirements/portable/requirements_cuda131.txt
+++ b/requirements/portable/requirements_cuda131.txt
@@ -23,5 +23,5 @@ sse-starlette==1.6.5
tiktoken
# CUDA wheels
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+cu131-py3-none-win_amd64.whl; platform_system == "Windows"
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+cu131-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+cu131-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+cu131-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
diff --git a/requirements/portable/requirements_vulkan.txt b/requirements/portable/requirements_vulkan.txt
index 76bb5872..dad7ee9f 100644
--- a/requirements/portable/requirements_vulkan.txt
+++ b/requirements/portable/requirements_vulkan.txt
@@ -23,5 +23,5 @@ sse-starlette==1.6.5
tiktoken
# Vulkan wheels
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+vulkan-py3-none-win_amd64.whl; platform_system == "Windows"
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.91.0/llama_cpp_binaries-0.91.0+vulkan-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+vulkan-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+vulkan-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
From e0a38da9f31c95332a5ca863217b1b2e485aecdc Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Fri, 13 Mar 2026 11:00:12 -0300
Subject: [PATCH 068/116] Improve tool call parsing for Devstral/GPT-OSS and
preserve thinking across tool turns
---
extensions/openai/utils.py | 51 +++++++++++++++++++++++++++++++++++---
modules/chat.py | 16 ++++++++++--
2 files changed, 62 insertions(+), 5 deletions(-)
diff --git a/extensions/openai/utils.py b/extensions/openai/utils.py
index eb34ce88..b179c267 100644
--- a/extensions/openai/utils.py
+++ b/extensions/openai/utils.py
@@ -126,8 +126,49 @@ def _parseChannelToolCalls(answer: str, tool_names: list[str]):
"""
matches = []
start_pos = None
- for m in re.finditer(
+ # Pattern 1: to=functions.NAME before <|channel|> (GPT-OSS primary format)
+ # Pattern 2: to=functions.NAME after <|channel|> (alternative format)
+ patterns = [
+ r'to=functions\.([^<\s]+)\s*<\|channel\|>[^<]*<\|message\|>',
r'<\|channel\|>\w+ to=functions\.([^<\s]+).*?<\|message\|>',
+ ]
+ for pattern in patterns:
+ for m in re.finditer(pattern, answer):
+ func_name = m.group(1).strip()
+ if func_name not in tool_names:
+ continue
+ json_str = _extractBalancedJson(answer, m.end())
+ if json_str is None:
+ continue
+ try:
+ arguments = json.loads(json_str)
+ if start_pos is None:
+ prefix = answer.rfind('<|start|>assistant', 0, m.start())
+ start_pos = prefix if prefix != -1 else m.start()
+ matches.append({
+ "type": "function",
+ "function": {
+ "name": func_name,
+ "arguments": arguments
+ }
+ })
+ except json.JSONDecodeError:
+ pass
+ if matches:
+ break
+ return matches, start_pos
+
+
+def _parseMistralTokenToolCalls(answer: str, tool_names: list[str]):
+ """Parse Mistral/Devstral-style tool calls with [TOOL_CALLS] and [ARGS] special tokens.
+
+ Format:
+ [TOOL_CALLS]func_name[ARGS]{"arg": "value"}
+ """
+ matches = []
+ start_pos = None
+ for m in re.finditer(
+ r'\[TOOL_CALLS\]\s*(\S+?)\s*\[ARGS\]\s*',
answer
):
func_name = m.group(1).strip()
@@ -139,8 +180,7 @@ def _parseChannelToolCalls(answer: str, tool_names: list[str]):
try:
arguments = json.loads(json_str)
if start_pos is None:
- prefix = answer.rfind('<|start|>assistant', 0, m.start())
- start_pos = prefix if prefix != -1 else m.start()
+ start_pos = m.start()
matches.append({
"type": "function",
"function": {
@@ -497,6 +537,11 @@ def parseToolCall(answer: str, tool_names: list[str], return_prefix: bool = Fals
if matches:
return _return(matches, start_pos)
+ # Check for Mistral/Devstral-style tool calls ([TOOL_CALLS]name[ARGS]json)
+ matches, start_pos = _parseMistralTokenToolCalls(answer, tool_names)
+ if matches:
+ return _return(matches, start_pos)
+
# Check for bare function-name style tool calls (e.g. Mistral format)
matches, start_pos = _parseBareNameToolCalls(answer, tool_names)
if matches:
diff --git a/modules/chat.py b/modules/chat.py
index 57fd50e0..2c6f0ab2 100644
--- a/modules/chat.py
+++ b/modules/chat.py
@@ -177,7 +177,7 @@ def _expand_tool_sequence(tool_seq):
deserialized = _deserialize_tool_call_arguments(item['tool_calls'])
messages.append({
"role": "assistant",
- "content": "",
+ "content": item.get('content', ''),
"tool_calls": deserialized
})
for tc in item['tool_calls']:
@@ -1324,7 +1324,19 @@ def generate_chat_reply_wrapper(text, state, regenerate=False, _continue=False):
tc_headers.append(f'{fn_name}({args_summary})')
- seq.append({'tool_calls': serialized})
+ seq_entry = {'tool_calls': serialized}
+ if content_prefix.strip():
+ # Strip GPT-OSS channel tokens so they don't get double-wrapped
+ # by the template (which adds its own channel markup).
+ clean = content_prefix.strip()
+ if '<|channel|>' in clean and '<|message|>' in clean:
+ inner = clean.split('<|message|>', 1)[1] if '<|message|>' in clean else clean
+ if '<|end|>' in inner:
+ inner = inner.split('<|end|>', 1)[0]
+ clean = inner.strip()
+ if clean:
+ seq_entry['content'] = clean
+ seq.append(seq_entry)
# Clear internal (raw tool markup)
history['internal'][-1][1] = ''
From aab2596d29a850a185eb0c25c83f7dcf7387d9fc Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Fri, 13 Mar 2026 15:47:11 -0300
Subject: [PATCH 069/116] UI: Fix multiple thinking blocks rendering as raw
text in HTML generator
---
modules/html_generator.py | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/modules/html_generator.py b/modules/html_generator.py
index 138d4ade..8f3f261f 100644
--- a/modules/html_generator.py
+++ b/modules/html_generator.py
@@ -369,8 +369,10 @@ def convert_to_markdown(string, message_id=None):
if not text.strip():
return
- thinking_content, remaining = extract_thinking_block(text)
- if thinking_content is not None:
+ while text.strip():
+ thinking_content, remaining = extract_thinking_block(text)
+ if thinking_content is None:
+ break
has_remaining = bool(remaining.strip()) or not is_last_segment
html_parts.append(build_thinking_block(thinking_content, message_id, has_remaining, think_idx))
think_idx += 1
From d4c22ced83dd6fed7bb2785c7e6d5f7cbeace2db Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Fri, 13 Mar 2026 15:47:14 -0300
Subject: [PATCH 070/116] UI: Optimize syntax highlighting and autoscroll by
moving from MutationObserver to morphdom updates
---
js/global_scope_js.js | 45 ++++++++++++++++
js/main.js | 121 ++++++++++++++----------------------------
2 files changed, 84 insertions(+), 82 deletions(-)
diff --git a/js/global_scope_js.js b/js/global_scope_js.js
index 3207a681..425c2c59 100644
--- a/js/global_scope_js.js
+++ b/js/global_scope_js.js
@@ -269,6 +269,34 @@ function removeLastClick() {
document.getElementById("Remove-last").click();
}
+function autoScrollToBottom() {
+ if (!window.isScrolled) {
+ const chatParent = document.getElementById("chat")?.parentNode?.parentNode?.parentNode;
+ if (chatParent) {
+ const maxScroll = chatParent.scrollHeight - chatParent.clientHeight;
+ if (maxScroll > 0 && chatParent.scrollTop < maxScroll - 1) {
+ chatParent.scrollTop = maxScroll;
+ }
+ }
+ }
+}
+
+function updateInstructPadding() {
+ const chatElement = document.getElementById("chat");
+ if (chatElement && chatElement.getAttribute("data-mode") === "instruct") {
+ const messagesContainer = chatElement.querySelector(".messages");
+ const lastChild = messagesContainer?.lastElementChild;
+ const prevSibling = lastChild?.previousElementSibling;
+ if (lastChild && prevSibling) {
+ let bufferHeight = Math.max(0, Math.max(window.innerHeight - 128 - 84, window.innerHeight - prevSibling.offsetHeight - 84) - lastChild.offsetHeight);
+ if (window.innerWidth <= 924) {
+ bufferHeight = Math.max(0, bufferHeight - 32);
+ }
+ messagesContainer.style.paddingBottom = `${bufferHeight}px`;
+ }
+ }
+}
+
let pendingMorphdomData = null;
let morphdomRafId = null;
@@ -373,10 +401,23 @@ function applyMorphdomUpdate(data) {
}
);
+ // Syntax highlighting and LaTeX
+ if (window.doSyntaxHighlighting) {
+ window.doSyntaxHighlighting();
+ }
+
+ // Auto-scroll runs both before and after padding update.
+ // Before: so content growth isn't hidden by padding absorption.
+ // After: so padding-added space is also scrolled into view.
+ autoScrollToBottom();
+ updateInstructPadding();
+ autoScrollToBottom();
+
// Add toggle listeners for new blocks
queryScope.querySelectorAll(".thinking-block").forEach(block => {
if (!block._hasToggleListener) {
block.addEventListener("toggle", function(e) {
+ const wasScrolled = window.isScrolled;
if (this.open) {
const content = this.querySelector(".thinking-content");
if (content) {
@@ -385,6 +426,10 @@ function applyMorphdomUpdate(data) {
}, 0);
}
}
+ updateInstructPadding();
+ // Restore scroll state so the browser's layout adjustment
+ // from the toggle doesn't disable auto-scroll
+ window.isScrolled = wasScrolled;
});
block._hasToggleListener = true;
}
diff --git a/js/main.js b/js/main.js
index c3e51c3c..0cefaa6e 100644
--- a/js/main.js
+++ b/js/main.js
@@ -147,6 +147,7 @@ window.isScrolled = false;
let scrollTimeout;
let lastScrollTop = 0;
let lastScrollHeight = 0;
+let lastClientHeight = 0;
targetElement.addEventListener("scroll", function() {
let diff = targetElement.scrollHeight - targetElement.clientHeight;
@@ -159,11 +160,12 @@ targetElement.addEventListener("scroll", function() {
if(isAtBottomNow) {
window.isScrolled = false;
- } else if (targetElement.scrollTop < lastScrollTop && targetElement.scrollHeight >= lastScrollHeight) {
+ } else if (targetElement.scrollTop < lastScrollTop && targetElement.scrollHeight >= lastScrollHeight && targetElement.clientHeight <= lastClientHeight) {
window.isScrolled = true;
}
lastScrollTop = targetElement.scrollTop;
lastScrollHeight = targetElement.scrollHeight;
+ lastClientHeight = targetElement.clientHeight;
// Clear previous timeout and set new one
clearTimeout(scrollTimeout);
@@ -174,14 +176,7 @@ targetElement.addEventListener("scroll", function() {
});
// Create a MutationObserver instance
-const observer = new MutationObserver(function(mutations) {
- // Check if this is just the scrolling class being toggled
- const isScrollingClassOnly = mutations.every(mutation =>
- mutation.type === "attributes" &&
- mutation.attributeName === "class" &&
- mutation.target === targetElement
- );
-
+const observer = new MutationObserver(function() {
if (targetElement.classList.contains("_generating")) {
typing.parentNode.classList.add("visible-dots");
document.getElementById("stop").style.display = "flex";
@@ -191,44 +186,11 @@ const observer = new MutationObserver(function(mutations) {
document.getElementById("stop").style.display = "none";
document.getElementById("Generate").style.display = "flex";
}
-
- doSyntaxHighlighting();
-
- if (!window.isScrolled && !isScrollingClassOnly) {
- const maxScroll = targetElement.scrollHeight - targetElement.clientHeight;
- if (maxScroll > 0 && targetElement.scrollTop < maxScroll - 1) {
- targetElement.scrollTop = maxScroll;
- }
- }
-
- const chatElement = document.getElementById("chat");
- if (chatElement && chatElement.getAttribute("data-mode") === "instruct") {
- const messagesContainer = chatElement.querySelector(".messages");
- const lastChild = messagesContainer?.lastElementChild;
- const prevSibling = lastChild?.previousElementSibling;
- if (lastChild && prevSibling) {
- // Add padding to the messages container to create room for the last message.
- // The purpose of this is to avoid constant scrolling during streaming in
- // instruct mode.
- let bufferHeight = Math.max(0, Math.max(window.innerHeight - 128 - 84, window.innerHeight - prevSibling.offsetHeight - 84) - lastChild.offsetHeight);
-
- // Subtract header height when screen width is <= 924px
- if (window.innerWidth <= 924) {
- bufferHeight = Math.max(0, bufferHeight - 32);
- }
-
- messagesContainer.style.paddingBottom = `${bufferHeight}px`;
- }
- }
});
-// Configure the observer to watch for changes in the subtree and attributes
+// Only watch for attribute changes on targetElement (e.g. _generating class)
const config = {
- childList: true,
- subtree: true,
- characterData: true,
- attributeOldValue: true,
- characterDataOldValue: true
+ attributes: true
};
// Start observing the target element
@@ -247,55 +209,50 @@ function isElementVisibleOnScreen(element) {
);
}
-function doSyntaxHighlighting() {
+window.doSyntaxHighlighting = function() {
const messageBodies = document.getElementById("chat").querySelectorAll(".message-body");
if (messageBodies.length > 0) {
- observer.disconnect();
+ let hasSeenVisible = false;
- try {
- let hasSeenVisible = false;
+ // Go from last message to first
+ for (let i = messageBodies.length - 1; i >= 0; i--) {
+ const messageBody = messageBodies[i];
- // Go from last message to first
- for (let i = messageBodies.length - 1; i >= 0; i--) {
- const messageBody = messageBodies[i];
+ if (isElementVisibleOnScreen(messageBody)) {
+ hasSeenVisible = true;
- if (isElementVisibleOnScreen(messageBody)) {
- hasSeenVisible = true;
+ // Handle both code and math in a single pass through each message
+ const codeBlocks = messageBody.querySelectorAll("pre code:not([data-highlighted])");
+ codeBlocks.forEach((codeBlock) => {
+ hljs.highlightElement(codeBlock);
+ codeBlock.setAttribute("data-highlighted", "true");
+ codeBlock.classList.add("pretty_scrollbar");
+ });
- // Handle both code and math in a single pass through each message
- const codeBlocks = messageBody.querySelectorAll("pre code:not([data-highlighted])");
- codeBlocks.forEach((codeBlock) => {
- hljs.highlightElement(codeBlock);
- codeBlock.setAttribute("data-highlighted", "true");
- codeBlock.classList.add("pretty_scrollbar");
- });
-
- // Only render math in visible elements
- const mathContainers = messageBody.querySelectorAll("p, span, li, td, th, h1, h2, h3, h4, h5, h6, blockquote, figcaption, caption, dd, dt");
- mathContainers.forEach(container => {
- if (isElementVisibleOnScreen(container)) {
- renderMathInElement(container, {
- delimiters: [
- { left: "$$", right: "$$", display: true },
- { left: "$", right: "$", display: false },
- { left: "\\(", right: "\\)", display: false },
- { left: "\\[", right: "\\]", display: true },
- ],
- });
- }
- });
- } else if (hasSeenVisible) {
- // We've seen visible messages but this one is not visible
- // Since we're going from last to first, we can break
- break;
- }
+ // Only render math in visible elements
+ const mathContainers = messageBody.querySelectorAll("p, span, li, td, th, h1, h2, h3, h4, h5, h6, blockquote, figcaption, caption, dd, dt");
+ mathContainers.forEach(container => {
+ if (isElementVisibleOnScreen(container)) {
+ renderMathInElement(container, {
+ delimiters: [
+ { left: "$$", right: "$$", display: true },
+ { left: "$", right: "$", display: false },
+ { left: "\\(", right: "\\)", display: false },
+ { left: "\\[", right: "\\]", display: true },
+ ],
+ });
+ }
+ });
+ } else if (hasSeenVisible) {
+ // We've seen visible messages but this one is not visible
+ // Since we're going from last to first, we can break
+ break;
}
- } finally {
- observer.observe(targetElement, config);
}
}
}
+const doSyntaxHighlighting = window.doSyntaxHighlighting;
//------------------------------------------------
// Add some scrollbars
From 5362bbb4132ae5ddbed4c4dab739e7dd64c1e6ab Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Fri, 13 Mar 2026 12:09:08 -0700
Subject: [PATCH 071/116] Make web_search not download the page contents, use
fetch_webpage instead
---
modules/ui_chat.py | 10 ++++++++++
modules/web_search.py | 14 ++++++++++++--
user_data/tools/web_search.py | 14 ++++++--------
3 files changed, 28 insertions(+), 10 deletions(-)
diff --git a/modules/ui_chat.py b/modules/ui_chat.py
index 039b9af6..ea341fa6 100644
--- a/modules/ui_chat.py
+++ b/modules/ui_chat.py
@@ -96,6 +96,16 @@ def create_ui():
shared.gradio['tools_refresh'] = gr.Button('Refresh list', elem_id='tools-refresh-btn', visible=False)
shared.gradio['tools_refresh'].click(fn=lambda: gr.update(choices=get_available_tools()), inputs=[], outputs=[shared.gradio['selected_tools']])
+ def sync_web_tools(selected):
+ if 'web_search' in selected and 'fetch_webpage' not in selected:
+ selected.append('fetch_webpage')
+ elif 'web_search' not in selected and 'fetch_webpage' in selected:
+ selected.remove('fetch_webpage')
+
+ return gr.update(value=selected)
+
+ shared.gradio['selected_tools'].change(fn=sync_web_tools, inputs=[shared.gradio['selected_tools']], outputs=[shared.gradio['selected_tools']], show_progress=False)
+
gr.HTML("")
with gr.Row():
diff --git a/modules/web_search.py b/modules/web_search.py
index b14cd042..754dd111 100644
--- a/modules/web_search.py
+++ b/modules/web_search.py
@@ -49,8 +49,8 @@ def download_web_page(url, timeout=10, include_links=False):
return ""
-def perform_web_search(query, num_pages=3, max_workers=5, timeout=10):
- """Perform web search and return results with content"""
+def perform_web_search(query, num_pages=3, max_workers=5, timeout=10, fetch_content=True):
+ """Perform web search and return results, optionally with page content"""
try:
search_url = f"https://html.duckduckgo.com/html/?q={quote_plus(query)}"
@@ -78,6 +78,16 @@ def perform_web_search(query, num_pages=3, max_workers=5, timeout=10):
search_results = [None] * len(download_tasks) # Pre-allocate to maintain order
+ if not fetch_content:
+ for url, title, index in download_tasks:
+ search_results[index] = {
+ 'title': title,
+ 'url': url,
+ 'content': ''
+ }
+
+ return search_results
+
# Download pages in parallel
with concurrent.futures.ThreadPoolExecutor(max_workers=max_workers) as executor:
# Submit all download tasks
diff --git a/user_data/tools/web_search.py b/user_data/tools/web_search.py
index 80845963..30d13473 100644
--- a/user_data/tools/web_search.py
+++ b/user_data/tools/web_search.py
@@ -1,16 +1,15 @@
-from modules.web_search import perform_web_search, truncate_content_by_tokens
+from modules.web_search import perform_web_search
tool = {
"type": "function",
"function": {
"name": "web_search",
- "description": "Search the web using DuckDuckGo and return page contents.",
+ "description": "Search the web using DuckDuckGo and return a list of result titles and URLs. Use fetch_webpage to read the contents of a specific result.",
"parameters": {
"type": "object",
"properties": {
"query": {"type": "string", "description": "The search query."},
- "num_pages": {"type": "integer", "description": "Number of search result pages to fetch (default: 3)."},
- "max_tokens": {"type": "integer", "description": "Maximum number of tokens per page result (default: 2048)."},
+ "num_pages": {"type": "integer", "description": "Number of search results to return (default: 3)."},
},
"required": ["query"]
}
@@ -21,11 +20,10 @@ tool = {
def execute(arguments):
query = arguments.get("query", "")
num_pages = arguments.get("num_pages", 3)
- max_tokens = arguments.get("max_tokens", 2048)
- results = perform_web_search(query, num_pages=num_pages)
+ results = perform_web_search(query, num_pages=num_pages, fetch_content=False)
output = []
for r in results:
- if r and r["content"].strip():
- output.append({"title": r["title"], "url": r["url"], "content": truncate_content_by_tokens(r["content"], max_tokens=max_tokens)})
+ if r:
+ output.append({"title": r["title"], "url": r["url"]})
return output if output else [{"error": "No results found."}]
From cabb95f0d6077d44741eb2a3ee0587470ade6300 Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Fri, 13 Mar 2026 12:24:48 -0700
Subject: [PATCH 072/116] UI: Increase the instruct width to 768px
---
css/html_instruct_style.css | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/css/html_instruct_style.css b/css/html_instruct_style.css
index 72a148c3..d4780350 100644
--- a/css/html_instruct_style.css
+++ b/css/html_instruct_style.css
@@ -78,7 +78,7 @@
.chat .user-message .text,
.chat .assistant-message .text {
- max-width: 700px;
+ max-width: 768px;
margin-left: auto;
margin-right: auto;
}
From 24e7e77b55e7758f7bd07e07016cc88b8b188c8b Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Fri, 13 Mar 2026 12:37:10 -0700
Subject: [PATCH 073/116] Clean up
---
modules/chat.py | 2 +-
modules/ui_chat.py | 2 --
2 files changed, 1 insertion(+), 3 deletions(-)
diff --git a/modules/chat.py b/modules/chat.py
index 2c6f0ab2..87e52851 100644
--- a/modules/chat.py
+++ b/modules/chat.py
@@ -1330,7 +1330,7 @@ def generate_chat_reply_wrapper(text, state, regenerate=False, _continue=False):
# by the template (which adds its own channel markup).
clean = content_prefix.strip()
if '<|channel|>' in clean and '<|message|>' in clean:
- inner = clean.split('<|message|>', 1)[1] if '<|message|>' in clean else clean
+ inner = clean.split('<|message|>', 1)[1]
if '<|end|>' in inner:
inner = inner.split('<|end|>', 1)[0]
clean = inner.strip()
diff --git a/modules/ui_chat.py b/modules/ui_chat.py
index ea341fa6..ce9fc0a2 100644
--- a/modules/ui_chat.py
+++ b/modules/ui_chat.py
@@ -99,8 +99,6 @@ def create_ui():
def sync_web_tools(selected):
if 'web_search' in selected and 'fetch_webpage' not in selected:
selected.append('fetch_webpage')
- elif 'web_search' not in selected and 'fetch_webpage' in selected:
- selected.remove('fetch_webpage')
return gr.update(value=selected)
From 0cd245bcbb46d894989e1ddd688d6640ba6ac537 Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Fri, 13 Mar 2026 12:58:56 -0700
Subject: [PATCH 074/116] UI: Make autoscroll more robust after the
optimizations
---
js/main.js | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/js/main.js b/js/main.js
index 0cefaa6e..a8bbbc71 100644
--- a/js/main.js
+++ b/js/main.js
@@ -181,6 +181,13 @@ const observer = new MutationObserver(function() {
typing.parentNode.classList.add("visible-dots");
document.getElementById("stop").style.display = "flex";
document.getElementById("Generate").style.display = "none";
+ // If the user is near the bottom, ensure auto-scroll is enabled
+ // for the new reply. This catches cases where isScrolled was
+ // incorrectly set to true by layout shifts during page load, etc.
+ const diff = targetElement.scrollHeight - targetElement.clientHeight;
+ if (Math.abs(targetElement.scrollTop - diff) <= 10 || diff <= 0) {
+ window.isScrolled = false;
+ }
} else {
typing.parentNode.classList.remove("visible-dots");
document.getElementById("stop").style.display = "none";
From cb88066d15c6ffa54774f805190f000951c05e84 Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Fri, 13 Mar 2026 13:17:41 -0700
Subject: [PATCH 075/116] Update llama.cpp
---
requirements/full/requirements.txt | 4 ++--
requirements/full/requirements_amd.txt | 4 ++--
requirements/full/requirements_apple_intel.txt | 2 +-
requirements/full/requirements_apple_silicon.txt | 2 +-
requirements/full/requirements_cpu_only.txt | 4 ++--
requirements/portable/requirements.txt | 4 ++--
requirements/portable/requirements_amd.txt | 4 ++--
requirements/portable/requirements_apple_intel.txt | 2 +-
requirements/portable/requirements_apple_silicon.txt | 2 +-
requirements/portable/requirements_cpu_only.txt | 4 ++--
requirements/portable/requirements_cuda131.txt | 4 ++--
requirements/portable/requirements_vulkan.txt | 4 ++--
12 files changed, 20 insertions(+), 20 deletions(-)
diff --git a/requirements/full/requirements.txt b/requirements/full/requirements.txt
index c702a8d3..8a0802f7 100644
--- a/requirements/full/requirements.txt
+++ b/requirements/full/requirements.txt
@@ -40,8 +40,8 @@ sse-starlette==1.6.5
tiktoken
# CUDA wheels
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+cu124-py3-none-win_amd64.whl; platform_system == "Windows"
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+cu124-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.93.0/llama_cpp_binaries-0.93.0+cu124-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.93.0/llama_cpp_binaries-0.93.0+cu124-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
https://github.com/turboderp-org/exllamav3/releases/download/v0.0.25/exllamav3-0.0.25+cu128.torch2.9.0-cp313-cp313-win_amd64.whl; platform_system == "Windows" and python_version == "3.13"
https://github.com/turboderp-org/exllamav3/releases/download/v0.0.25/exllamav3-0.0.25+cu128.torch2.9.0-cp313-cp313-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.13"
https://github.com/kingbri1/flash-attention/releases/download/v2.8.3/flash_attn-2.8.3+cu128torch2.9.0cxx11abiFALSE-cp313-cp313-win_amd64.whl; platform_system == "Windows" and python_version == "3.13"
diff --git a/requirements/full/requirements_amd.txt b/requirements/full/requirements_amd.txt
index 65a9aa00..9b31d668 100644
--- a/requirements/full/requirements_amd.txt
+++ b/requirements/full/requirements_amd.txt
@@ -37,5 +37,5 @@ sse-starlette==1.6.5
tiktoken
# AMD wheels
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+rocm7.2-py3-none-win_amd64.whl; platform_system == "Windows"
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+rocm7.2-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.93.0/llama_cpp_binaries-0.93.0+rocm7.2-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.93.0/llama_cpp_binaries-0.93.0+rocm7.2-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
diff --git a/requirements/full/requirements_apple_intel.txt b/requirements/full/requirements_apple_intel.txt
index bba62491..138639e5 100644
--- a/requirements/full/requirements_apple_intel.txt
+++ b/requirements/full/requirements_apple_intel.txt
@@ -37,4 +37,4 @@ sse-starlette==1.6.5
tiktoken
# Mac wheels
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0-py3-none-macosx_13_0_x86_64.whl; platform_system == "Darwin"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.93.0/llama_cpp_binaries-0.93.0-py3-none-macosx_13_0_x86_64.whl; platform_system == "Darwin"
diff --git a/requirements/full/requirements_apple_silicon.txt b/requirements/full/requirements_apple_silicon.txt
index 61dbf51b..f3ebd171 100644
--- a/requirements/full/requirements_apple_silicon.txt
+++ b/requirements/full/requirements_apple_silicon.txt
@@ -37,4 +37,4 @@ sse-starlette==1.6.5
tiktoken
# Mac wheels
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0-py3-none-macosx_13_0_arm64.whl; platform_system == "Darwin"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.93.0/llama_cpp_binaries-0.93.0-py3-none-macosx_13_0_arm64.whl; platform_system == "Darwin"
diff --git a/requirements/full/requirements_cpu_only.txt b/requirements/full/requirements_cpu_only.txt
index 384a552a..e32a2ed1 100644
--- a/requirements/full/requirements_cpu_only.txt
+++ b/requirements/full/requirements_cpu_only.txt
@@ -37,5 +37,5 @@ sse-starlette==1.6.5
tiktoken
# llama.cpp (CPU only)
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+cpu-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+cpu-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.93.0/llama_cpp_binaries-0.93.0+cpu-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.93.0/llama_cpp_binaries-0.93.0+cpu-py3-none-win_amd64.whl; platform_system == "Windows"
diff --git a/requirements/portable/requirements.txt b/requirements/portable/requirements.txt
index 0e3d67d3..93eb3b85 100644
--- a/requirements/portable/requirements.txt
+++ b/requirements/portable/requirements.txt
@@ -23,5 +23,5 @@ sse-starlette==1.6.5
tiktoken
# CUDA wheels
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+cu124-py3-none-win_amd64.whl; platform_system == "Windows"
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+cu124-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.93.0/llama_cpp_binaries-0.93.0+cu124-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.93.0/llama_cpp_binaries-0.93.0+cu124-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
diff --git a/requirements/portable/requirements_amd.txt b/requirements/portable/requirements_amd.txt
index 729829b3..36e0e4d9 100644
--- a/requirements/portable/requirements_amd.txt
+++ b/requirements/portable/requirements_amd.txt
@@ -23,5 +23,5 @@ sse-starlette==1.6.5
tiktoken
# AMD wheels
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+rocm7.2-py3-none-win_amd64.whl; platform_system == "Windows"
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+rocm7.2-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.93.0/llama_cpp_binaries-0.93.0+rocm7.2-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.93.0/llama_cpp_binaries-0.93.0+rocm7.2-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
diff --git a/requirements/portable/requirements_apple_intel.txt b/requirements/portable/requirements_apple_intel.txt
index 4b16414c..495bd5fa 100644
--- a/requirements/portable/requirements_apple_intel.txt
+++ b/requirements/portable/requirements_apple_intel.txt
@@ -23,4 +23,4 @@ sse-starlette==1.6.5
tiktoken
# Mac wheels
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0-py3-none-macosx_13_0_x86_64.whl; platform_system == "Darwin"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.93.0/llama_cpp_binaries-0.93.0-py3-none-macosx_13_0_x86_64.whl; platform_system == "Darwin"
diff --git a/requirements/portable/requirements_apple_silicon.txt b/requirements/portable/requirements_apple_silicon.txt
index 3a1764dc..7e82f68d 100644
--- a/requirements/portable/requirements_apple_silicon.txt
+++ b/requirements/portable/requirements_apple_silicon.txt
@@ -23,4 +23,4 @@ sse-starlette==1.6.5
tiktoken
# Mac wheels
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0-py3-none-macosx_13_0_arm64.whl; platform_system == "Darwin"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.93.0/llama_cpp_binaries-0.93.0-py3-none-macosx_13_0_arm64.whl; platform_system == "Darwin"
diff --git a/requirements/portable/requirements_cpu_only.txt b/requirements/portable/requirements_cpu_only.txt
index 9d115c86..046619e1 100644
--- a/requirements/portable/requirements_cpu_only.txt
+++ b/requirements/portable/requirements_cpu_only.txt
@@ -23,5 +23,5 @@ sse-starlette==1.6.5
tiktoken
# llama.cpp (CPU only)
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+cpu-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+cpu-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.93.0/llama_cpp_binaries-0.93.0+cpu-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.93.0/llama_cpp_binaries-0.93.0+cpu-py3-none-win_amd64.whl; platform_system == "Windows"
diff --git a/requirements/portable/requirements_cuda131.txt b/requirements/portable/requirements_cuda131.txt
index 4472e1d4..590562f8 100644
--- a/requirements/portable/requirements_cuda131.txt
+++ b/requirements/portable/requirements_cuda131.txt
@@ -23,5 +23,5 @@ sse-starlette==1.6.5
tiktoken
# CUDA wheels
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+cu131-py3-none-win_amd64.whl; platform_system == "Windows"
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+cu131-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.93.0/llama_cpp_binaries-0.93.0+cu131-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.93.0/llama_cpp_binaries-0.93.0+cu131-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
diff --git a/requirements/portable/requirements_vulkan.txt b/requirements/portable/requirements_vulkan.txt
index dad7ee9f..bf80deb0 100644
--- a/requirements/portable/requirements_vulkan.txt
+++ b/requirements/portable/requirements_vulkan.txt
@@ -23,5 +23,5 @@ sse-starlette==1.6.5
tiktoken
# Vulkan wheels
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+vulkan-py3-none-win_amd64.whl; platform_system == "Windows"
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.92.0/llama_cpp_binaries-0.92.0+vulkan-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.93.0/llama_cpp_binaries-0.93.0+vulkan-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.93.0/llama_cpp_binaries-0.93.0+vulkan-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
From e8d1c663037666bafc0a45f4be0471a88fda4d57 Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Fri, 13 Mar 2026 18:13:12 -0700
Subject: [PATCH 076/116] Clean up tool calling code
---
extensions/openai/completions.py | 7 +-
extensions/openai/utils.py | 558 -------------------------------
modules/chat.py | 17 +-
modules/tool_parsing.py | 553 ++++++++++++++++++++++++++++++
modules/tool_use.py | 6 +-
modules/ui_chat.py | 2 +-
modules/web_search.py | 37 +-
7 files changed, 604 insertions(+), 576 deletions(-)
create mode 100644 modules/tool_parsing.py
diff --git a/extensions/openai/completions.py b/extensions/openai/completions.py
index 290a5bc0..27defe42 100644
--- a/extensions/openai/completions.py
+++ b/extensions/openai/completions.py
@@ -11,7 +11,8 @@ from pydantic import ValidationError
from extensions.openai.errors import InvalidRequestError
from extensions.openai.typing import ToolDefinition
-from extensions.openai.utils import debug_msg, getToolCallId, parseToolCall
+from extensions.openai.utils import debug_msg
+from modules.tool_parsing import get_tool_call_id, parse_tool_call
from modules import shared
from modules.reasoning import extract_reasoning
from modules.chat import (
@@ -491,10 +492,10 @@ def chat_completions_common(body: dict, is_legacy: bool = False, stream=False, p
answer = a['internal'][-1][1]
if supported_tools is not None:
- tool_call = parseToolCall(answer[end_last_tool_call:], supported_tools) if len(answer) > 0 else []
+ tool_call = parse_tool_call(answer[end_last_tool_call:], supported_tools) if len(answer) > 0 else []
if len(tool_call) > 0:
for tc in tool_call:
- tc["id"] = getToolCallId()
+ tc["id"] = get_tool_call_id()
if stream:
tc["index"] = len(tool_calls)
tc["function"]["arguments"] = json.dumps(tc["function"]["arguments"])
diff --git a/extensions/openai/utils.py b/extensions/openai/utils.py
index b179c267..2b414769 100644
--- a/extensions/openai/utils.py
+++ b/extensions/openai/utils.py
@@ -1,8 +1,5 @@
import base64
-import json
import os
-import random
-import re
import time
import traceback
from typing import Callable, Optional
@@ -55,558 +52,3 @@ def _start_cloudflared(port: int, tunnel_id: str, max_attempts: int = 3, on_star
time.sleep(3)
raise Exception('Could not start cloudflared.')
-
-
-def getToolCallId() -> str:
- letter_bytes = "abcdefghijklmnopqrstuvwxyz0123456789"
- b = [random.choice(letter_bytes) for _ in range(8)]
- return "call_" + "".join(b).lower()
-
-
-def checkAndSanitizeToolCallCandidate(candidate_dict: dict, tool_names: list[str]):
- # check if property 'function' exists and is a dictionary, otherwise adapt dict
- if 'function' not in candidate_dict and 'name' in candidate_dict and isinstance(candidate_dict['name'], str):
- candidate_dict = {"type": "function", "function": candidate_dict}
- if 'function' in candidate_dict and isinstance(candidate_dict['function'], str):
- candidate_dict['name'] = candidate_dict['function']
- del candidate_dict['function']
- candidate_dict = {"type": "function", "function": candidate_dict}
- if 'function' in candidate_dict and isinstance(candidate_dict['function'], dict):
- # check if 'name' exists within 'function' and is part of known tools
- if 'name' in candidate_dict['function'] and candidate_dict['function']['name'] in tool_names:
- candidate_dict["type"] = "function" # ensure required property 'type' exists and has the right value
- # map property 'parameters' used by some older models to 'arguments'
- if "arguments" not in candidate_dict["function"] and "parameters" in candidate_dict["function"]:
- candidate_dict["function"]["arguments"] = candidate_dict["function"]["parameters"]
- del candidate_dict["function"]["parameters"]
- return candidate_dict
- return None
-
-
-def _extractBalancedJson(text: str, start: int) -> str | None:
- """Extract a balanced JSON object from text starting at the given position.
-
- Walks through the string tracking brace depth and string boundaries
- to correctly handle arbitrary nesting levels.
- """
- if start >= len(text) or text[start] != '{':
- return None
- depth = 0
- in_string = False
- escape_next = False
- for i in range(start, len(text)):
- c = text[i]
- if escape_next:
- escape_next = False
- continue
- if c == '\\' and in_string:
- escape_next = True
- continue
- if c == '"':
- in_string = not in_string
- continue
- if in_string:
- continue
- if c == '{':
- depth += 1
- elif c == '}':
- depth -= 1
- if depth == 0:
- return text[start:i + 1]
- return None
-
-
-def _parseChannelToolCalls(answer: str, tool_names: list[str]):
- """Parse channel-based tool calls used by GPT-OSS and similar models.
-
- Format:
- <|start|>assistant to=functions.func_name<|channel|>commentary json<|message|>{"arg": "value"}
- or:
- <|channel|>commentary to=functions.func_name <|constrain|>json<|message|>{"arg": "value"}
- """
- matches = []
- start_pos = None
- # Pattern 1: to=functions.NAME before <|channel|> (GPT-OSS primary format)
- # Pattern 2: to=functions.NAME after <|channel|> (alternative format)
- patterns = [
- r'to=functions\.([^<\s]+)\s*<\|channel\|>[^<]*<\|message\|>',
- r'<\|channel\|>\w+ to=functions\.([^<\s]+).*?<\|message\|>',
- ]
- for pattern in patterns:
- for m in re.finditer(pattern, answer):
- func_name = m.group(1).strip()
- if func_name not in tool_names:
- continue
- json_str = _extractBalancedJson(answer, m.end())
- if json_str is None:
- continue
- try:
- arguments = json.loads(json_str)
- if start_pos is None:
- prefix = answer.rfind('<|start|>assistant', 0, m.start())
- start_pos = prefix if prefix != -1 else m.start()
- matches.append({
- "type": "function",
- "function": {
- "name": func_name,
- "arguments": arguments
- }
- })
- except json.JSONDecodeError:
- pass
- if matches:
- break
- return matches, start_pos
-
-
-def _parseMistralTokenToolCalls(answer: str, tool_names: list[str]):
- """Parse Mistral/Devstral-style tool calls with [TOOL_CALLS] and [ARGS] special tokens.
-
- Format:
- [TOOL_CALLS]func_name[ARGS]{"arg": "value"}
- """
- matches = []
- start_pos = None
- for m in re.finditer(
- r'\[TOOL_CALLS\]\s*(\S+?)\s*\[ARGS\]\s*',
- answer
- ):
- func_name = m.group(1).strip()
- if func_name not in tool_names:
- continue
- json_str = _extractBalancedJson(answer, m.end())
- if json_str is None:
- continue
- try:
- arguments = json.loads(json_str)
- if start_pos is None:
- start_pos = m.start()
- matches.append({
- "type": "function",
- "function": {
- "name": func_name,
- "arguments": arguments
- }
- })
- except json.JSONDecodeError:
- pass
- return matches, start_pos
-
-
-def _parseBareNameToolCalls(answer: str, tool_names: list[str]):
- """Parse bare function-name style tool calls used by Mistral and similar models.
-
- Format:
- functionName{"arg": "value"}
- Multiple calls are concatenated directly or separated by whitespace.
- """
- matches = []
- start_pos = None
- # Match tool name followed by opening brace, then extract balanced JSON
- escaped_names = [re.escape(name) for name in tool_names]
- pattern = r'(?:' + '|'.join(escaped_names) + r')\s*\{'
- for match in re.finditer(pattern, answer):
- text = match.group(0)
- name = None
- for n in tool_names:
- if text.startswith(n):
- name = n
- break
- if not name:
- continue
- brace_start = match.end() - 1
- json_str = _extractBalancedJson(answer, brace_start)
- if json_str is None:
- continue
- try:
- arguments = json.loads(json_str)
- if start_pos is None:
- start_pos = match.start()
- matches.append({
- "type": "function",
- "function": {
- "name": name,
- "arguments": arguments
- }
- })
- except json.JSONDecodeError:
- pass
- return matches, start_pos
-
-
-def _parseXmlParamToolCalls(answer: str, tool_names: list[str]):
- """Parse XML-parameter style tool calls used by Qwen3.5 and similar models.
-
- Format:
-
-
- value
-