oobabooga/text-generation-webui
1
0
Fork 0
mirror of https://github.com/oobabooga/text-generation-webui.git synced 2025-12-06 07:12:10 +01:00
Code Issues Actions Packages Projects Releases Wiki Activity

Merge pull request #7331 from oobabooga/dev

Browse source 
Merge dev branch
This commit is contained in:
oobabooga 2025-11-28 23:00:01 -03:00 committed by GitHub
parent 1afe0827ba 661e42d2b7
commit bd9f2de73a
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
26 changed files with 117 additions and 83 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
.github/workflows/build-portable-release.yml vendored
View file

@ -57,7 +57,7 @@ jobs:
id: set-matrix id: set-matrix
run: | run: |
$matrix = @{ $matrix = @{
'os' = @('ubuntu-22.04', 'windows-2022', 'macos-13', 'macos-14') 'os' = @('ubuntu-22.04', 'windows-2022', 'macos-14')
'pyver' = @("3.11") 'pyver' = @("3.11")
'avx' = @("AVX2") 'avx' = @("AVX2")
} }

2
extensions/coqui_tts/requirements.txt
View file

@ -1 +1 @@
coqui-tts==0.25.1 coqui-tts>=0.27.0

75
modules/html_generator.py
View file

@ -196,33 +196,22 @@ def extract_thinking_block(string):
return None, string return None, string
@functools.lru_cache(maxsize=None) def build_thinking_block(thinking_content, message_id, has_remaining_content):
def convert_to_markdown(string, message_id=None): """Build HTML for a thinking block."""
if not string: if thinking_content is None:
return "" return None
# Use a default message ID if none provided # Process the thinking content through markdown
if message_id is None:
message_id = "unknown"
# Extract thinking block if present
thinking_content, remaining_content = extract_thinking_block(string)
# Process the main content
html_output = process_markdown_content(remaining_content)
# If thinking content was found, process it using the same function
if thinking_content is not None:
thinking_html = process_markdown_content(thinking_content) thinking_html = process_markdown_content(thinking_content)
# Generate unique ID for the thinking block # Generate unique ID for the thinking block
block_id = f"thinking-{message_id}-0" block_id = f"thinking-{message_id}-0"
# Check if thinking is complete or still in progress # Check if thinking is complete or still in progress
is_streaming = not remaining_content is_streaming = not has_remaining_content
title_text = "Thinking..." if is_streaming else "Thought" title_text = "Thinking..." if is_streaming else "Thought"
thinking_block = f''' return f'''
<details class="thinking-block" data-block-id="{block_id}" data-streaming="{str(is_streaming).lower()}"> <details class="thinking-block" data-block-id="{block_id}" data-streaming="{str(is_streaming).lower()}">
<summary class="thinking-header"> <summary class="thinking-header">
{info_svg_small} {info_svg_small}
@ -232,14 +221,20 @@ def convert_to_markdown(string, message_id=None):
</details> </details>
''' '''
# Prepend the thinking block to the message HTML
html_output = thinking_block + html_output
return html_output def build_main_content_block(content):
"""Build HTML for the main content block."""
if not content:
return ""
return process_markdown_content(content)
def process_markdown_content(string): def process_markdown_content(string):
"""Process a string through the markdown conversion pipeline.""" """
Process a string through the markdown conversion pipeline.
Uses robust manual parsing to ensure correct LaTeX and Code Block rendering.
"""
if not string: if not string:
return "" return ""
@ -280,7 +275,7 @@ def process_markdown_content(string):
pattern = re.compile(r'\\begin{blockquote}(.*?)\\end{blockquote}', re.DOTALL) pattern = re.compile(r'\\begin{blockquote}(.*?)\\end{blockquote}', re.DOTALL)
string = pattern.sub(replace_blockquote, string) string = pattern.sub(replace_blockquote, string)
# Code # Code block standardization
string = string.replace('\\begin{code}', '```') string = string.replace('\\begin{code}', '```')
string = string.replace('\\end{code}', '```') string = string.replace('\\end{code}', '```')
string = string.replace('\\begin{align*}', '$$') string = string.replace('\\begin{align*}', '$$')
@ -301,6 +296,7 @@ def process_markdown_content(string):
is_code = False is_code = False
is_latex = False is_latex = False
# Manual line iteration for robust structure parsing
for line in string.split('\n'): for line in string.split('\n'):
stripped_line = line.strip() stripped_line = line.strip()
@ -371,6 +367,39 @@ def process_markdown_content(string):
return html_output return html_output
@functools.lru_cache(maxsize=None)
def convert_to_markdown(string, message_id=None):
"""
Convert a string to markdown HTML with support for multiple block types.
Blocks are assembled in order: thinking, main content, etc.
"""
if not string:
return ""
# Use a default message ID if none provided
if message_id is None:
message_id = "unknown"
# Extract different components from the string
thinking_content, remaining_content = extract_thinking_block(string)
# Build individual HTML blocks
blocks = []
# Add thinking block if present
thinking_html = build_thinking_block(thinking_content, message_id, bool(remaining_content))
if thinking_html:
blocks.append(thinking_html)
# Add main content block
main_html = build_main_content_block(remaining_content)
if main_html:
blocks.append(main_html)
# Assemble all blocks into final HTML
return ''.join(blocks)
def convert_to_markdown_wrapped(string, message_id=None, use_cache=True): def convert_to_markdown_wrapped(string, message_id=None, use_cache=True):
''' '''
Used to avoid caching convert_to_markdown calls during streaming. Used to avoid caching convert_to_markdown calls during streaming.

1
modules/llama_cpp_server.py
View file

@ -317,6 +317,7 @@ class LlamaServer:
"--ctx-size", str(shared.args.ctx_size), "--ctx-size", str(shared.args.ctx_size),
"--gpu-layers", str(shared.args.gpu_layers), "--gpu-layers", str(shared.args.gpu_layers),
"--batch-size", str(shared.args.batch_size), "--batch-size", str(shared.args.batch_size),
"--ubatch-size", str(shared.args.ubatch_size),
"--port", str(self.port), "--port", str(self.port),
"--no-webui", "--no-webui",
"--flash-attn", "on", "--flash-attn", "on",

1
modules/loaders.py
View file

@ -10,6 +10,7 @@ loaders_and_params = OrderedDict({
'threads', 'threads',
'threads_batch', 'threads_batch',
'batch_size', 'batch_size',
'ubatch_size',
'ctx_size', 'ctx_size',
'cache_type', 'cache_type',
'tensor_split', 'tensor_split',

3
modules/shared.py
View file

@ -74,7 +74,8 @@ group.add_argument('--row-split', action='store_true', help='Split the model by
group.add_argument('--no-mmap', action='store_true', help='Prevent mmap from being used.') group.add_argument('--no-mmap', action='store_true', help='Prevent mmap from being used.')
group.add_argument('--mlock', action='store_true', help='Force the system to keep the model in RAM.') group.add_argument('--mlock', action='store_true', help='Force the system to keep the model in RAM.')
group.add_argument('--no-kv-offload', action='store_true', help='Do not offload the K, Q, V to the GPU. This saves VRAM but reduces the performance.') group.add_argument('--no-kv-offload', action='store_true', help='Do not offload the K, Q, V to the GPU. This saves VRAM but reduces the performance.')
group.add_argument('--batch-size', type=int, default=256, help='Maximum number of prompt tokens to batch together when calling llama_eval.') group.add_argument('--batch-size', type=int, default=1024, help='Maximum number of prompt tokens to batch together when calling llama-server. This is the application level batch size.')
group.add_argument('--ubatch-size', type=int, default=1024, help='Maximum number of prompt tokens to batch together when calling llama-server. This is the max physical batch size for computation (device level).')
group.add_argument('--threads', type=int, default=0, help='Number of threads to use.') group.add_argument('--threads', type=int, default=0, help='Number of threads to use.')
group.add_argument('--threads-batch', type=int, default=0, help='Number of threads to use for batches/prompt processing.') group.add_argument('--threads-batch', type=int, default=0, help='Number of threads to use for batches/prompt processing.')
group.add_argument('--numa', action='store_true', help='Activate NUMA task allocation for llama.cpp.') group.add_argument('--numa', action='store_true', help='Activate NUMA task allocation for llama.cpp.')

1
modules/ui.py
View file

@ -129,6 +129,7 @@ def list_model_elements():
'threads', 'threads',
'threads_batch', 'threads_batch',
'batch_size', 'batch_size',
'ubatch_size',
'ctx_size', 'ctx_size',
'cache_type', 'cache_type',
'tensor_split', 'tensor_split',

1
modules/ui_model_menu.py
View file

@ -84,6 +84,7 @@ def create_ui():
shared.gradio['threads'] = gr.Slider(label="threads", minimum=0, step=1, maximum=256, value=shared.args.threads) shared.gradio['threads'] = gr.Slider(label="threads", minimum=0, step=1, maximum=256, value=shared.args.threads)
shared.gradio['threads_batch'] = gr.Slider(label="threads_batch", minimum=0, step=1, maximum=256, value=shared.args.threads_batch) shared.gradio['threads_batch'] = gr.Slider(label="threads_batch", minimum=0, step=1, maximum=256, value=shared.args.threads_batch)
shared.gradio['batch_size'] = gr.Slider(label="batch_size", minimum=1, maximum=4096, step=1, value=shared.args.batch_size) shared.gradio['batch_size'] = gr.Slider(label="batch_size", minimum=1, maximum=4096, step=1, value=shared.args.batch_size)
shared.gradio['ubatch_size'] = gr.Slider(label="ubatch_size", minimum=1, maximum=4096, step=1, value=shared.args.ubatch_size)
shared.gradio['tensor_split'] = gr.Textbox(label='tensor_split', info='List of proportions to split the model across multiple GPUs. Example: 60,40') shared.gradio['tensor_split'] = gr.Textbox(label='tensor_split', info='List of proportions to split the model across multiple GPUs. Example: 60,40')
shared.gradio['extra_flags'] = gr.Textbox(label='extra-flags', info='Additional flags to pass to llama-server. Format: "flag1=value1,flag2,flag3=value3". Example: "override-tensor=exps=CPU"', value=shared.args.extra_flags) shared.gradio['extra_flags'] = gr.Textbox(label='extra-flags', info='Additional flags to pass to llama-server. Format: "flag1=value1,flag2,flag3=value3". Example: "override-tensor=exps=CPU"', value=shared.args.extra_flags)
shared.gradio['cpu_memory'] = gr.Number(label="Maximum CPU memory in GiB. Use this for CPU offloading.", value=shared.args.cpu_memory) shared.gradio['cpu_memory'] = gr.Number(label="Maximum CPU memory in GiB. Use this for CPU offloading.", value=shared.args.cpu_memory)

8
requirements/full/requirements.txt
View file

@ -40,10 +40,10 @@ sse-starlette==1.6.5
tiktoken tiktoken
# CUDA wheels # CUDA wheels
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+cu124-py3-none-win_amd64.whl; platform_system == "Windows" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+cu124-py3-none-win_amd64.whl; platform_system == "Windows"
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+cu124-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+cu124-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
https://github.com/turboderp-org/exllamav3/releases/download/v0.0.15/exllamav3-0.0.15+cu128.torch2.7.0-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11" https://github.com/turboderp-org/exllamav3/releases/download/v0.0.16/exllamav3-0.0.16+cu128.torch2.7.0-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
https://github.com/turboderp-org/exllamav3/releases/download/v0.0.15/exllamav3-0.0.15+cu128.torch2.7.0-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11" https://github.com/turboderp-org/exllamav3/releases/download/v0.0.16/exllamav3-0.0.16+cu128.torch2.7.0-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
https://github.com/turboderp-org/exllamav2/releases/download/v0.3.2/exllamav2-0.3.2+cu128.torch2.7.0-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11" https://github.com/turboderp-org/exllamav2/releases/download/v0.3.2/exllamav2-0.3.2+cu128.torch2.7.0-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
https://github.com/turboderp-org/exllamav2/releases/download/v0.3.2/exllamav2-0.3.2+cu128.torch2.7.0-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11" https://github.com/turboderp-org/exllamav2/releases/download/v0.3.2/exllamav2-0.3.2+cu128.torch2.7.0-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
https://github.com/turboderp-org/exllamav2/releases/download/v0.3.2/exllamav2-0.3.2-py3-none-any.whl; platform_system == "Linux" and platform_machine != "x86_64" https://github.com/turboderp-org/exllamav2/releases/download/v0.3.2/exllamav2-0.3.2-py3-none-any.whl; platform_system == "Linux" and platform_machine != "x86_64"

4
requirements/full/requirements_amd.txt
View file

@ -38,7 +38,7 @@ sse-starlette==1.6.5
tiktoken tiktoken
# AMD wheels # AMD wheels
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+vulkan-py3-none-win_amd64.whl; platform_system == "Windows" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+vulkan-py3-none-win_amd64.whl; platform_system == "Windows"
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+rocm6.4.4-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+rocm6.4.4-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
https://github.com/turboderp-org/exllamav2/releases/download/v0.3.2/exllamav2-0.3.2+rocm6.2.4.torch2.6.0-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11" https://github.com/turboderp-org/exllamav2/releases/download/v0.3.2/exllamav2-0.3.2+rocm6.2.4.torch2.6.0-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
https://github.com/turboderp-org/exllamav2/releases/download/v0.3.2/exllamav2-0.3.2-py3-none-any.whl; platform_system != "Darwin" and platform_machine != "x86_64" https://github.com/turboderp-org/exllamav2/releases/download/v0.3.2/exllamav2-0.3.2-py3-none-any.whl; platform_system != "Darwin" and platform_machine != "x86_64"

4
requirements/full/requirements_amd_noavx2.txt
View file

@ -38,7 +38,7 @@ sse-starlette==1.6.5
tiktoken tiktoken
# AMD wheels # AMD wheels
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+vulkanavx-py3-none-win_amd64.whl; platform_system == "Windows" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+vulkanavx-py3-none-win_amd64.whl; platform_system == "Windows"
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+vulkanavx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+vulkanavx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
https://github.com/turboderp-org/exllamav2/releases/download/v0.3.2/exllamav2-0.3.2+rocm6.2.4.torch2.6.0-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11" https://github.com/turboderp-org/exllamav2/releases/download/v0.3.2/exllamav2-0.3.2+rocm6.2.4.torch2.6.0-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
https://github.com/turboderp-org/exllamav2/releases/download/v0.3.2/exllamav2-0.3.2-py3-none-any.whl; platform_system != "Darwin" and platform_machine != "x86_64" https://github.com/turboderp-org/exllamav2/releases/download/v0.3.2/exllamav2-0.3.2-py3-none-any.whl; platform_system != "Darwin" and platform_machine != "x86_64"

4
requirements/full/requirements_apple_intel.txt
View file

@ -38,5 +38,5 @@ sse-starlette==1.6.5
tiktoken tiktoken
# Mac wheels # Mac wheels
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0-py3-none-macosx_15_0_x86_64.whl; platform_system == "Darwin" and platform_release >= "24.0.0" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0-py3-none-macosx_15_0_x86_64.whl; platform_system == "Darwin" and platform_release >= "24.0.0"
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0-py3-none-macosx_14_0_x86_64.whl; platform_system == "Darwin" and platform_release >= "23.0.0" and platform_release < "24.0.0" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0-py3-none-macosx_14_0_x86_64.whl; platform_system == "Darwin" and platform_release >= "23.0.0" and platform_release < "24.0.0"

4
requirements/full/requirements_apple_silicon.txt
View file

@ -38,5 +38,5 @@ sse-starlette==1.6.5
tiktoken tiktoken
# Mac wheels # Mac wheels
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0-py3-none-macosx_15_0_arm64.whl; platform_system == "Darwin" and platform_release >= "24.0.0" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0-py3-none-macosx_15_0_arm64.whl; platform_system == "Darwin" and platform_release >= "24.0.0"
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0-py3-none-macosx_14_0_arm64.whl; platform_system == "Darwin" and platform_release >= "23.0.0" and platform_release < "24.0.0" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0-py3-none-macosx_14_0_arm64.whl; platform_system == "Darwin" and platform_release >= "23.0.0" and platform_release < "24.0.0"

4
requirements/full/requirements_cpu_only.txt
View file

@ -38,5 +38,5 @@ sse-starlette==1.6.5
tiktoken tiktoken
# llama.cpp (CPU only, AVX2) # llama.cpp (CPU only, AVX2)
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+cpuavx2-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+cpuavx2-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+cpuavx2-py3-none-win_amd64.whl; platform_system == "Windows" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+cpuavx2-py3-none-win_amd64.whl; platform_system == "Windows"

4
requirements/full/requirements_cpu_only_noavx2.txt
View file

@ -38,5 +38,5 @@ sse-starlette==1.6.5
tiktoken tiktoken
# llama.cpp (CPU only, no AVX2) # llama.cpp (CPU only, no AVX2)
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+cpuavx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+cpuavx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+cpuavx-py3-none-win_amd64.whl; platform_system == "Windows" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+cpuavx-py3-none-win_amd64.whl; platform_system == "Windows"

8
requirements/full/requirements_noavx2.txt
View file

@ -40,10 +40,10 @@ sse-starlette==1.6.5
tiktoken tiktoken
# CUDA wheels # CUDA wheels
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+cu124avx-py3-none-win_amd64.whl; platform_system == "Windows" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+cu124avx-py3-none-win_amd64.whl; platform_system == "Windows"
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+cu124avx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+cu124avx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
https://github.com/turboderp-org/exllamav3/releases/download/v0.0.15/exllamav3-0.0.15+cu128.torch2.7.0-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11" https://github.com/turboderp-org/exllamav3/releases/download/v0.0.16/exllamav3-0.0.16+cu128.torch2.7.0-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
https://github.com/turboderp-org/exllamav3/releases/download/v0.0.15/exllamav3-0.0.15+cu128.torch2.7.0-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11" https://github.com/turboderp-org/exllamav3/releases/download/v0.0.16/exllamav3-0.0.16+cu128.torch2.7.0-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
https://github.com/turboderp-org/exllamav2/releases/download/v0.3.2/exllamav2-0.3.2+cu128.torch2.7.0-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11" https://github.com/turboderp-org/exllamav2/releases/download/v0.3.2/exllamav2-0.3.2+cu128.torch2.7.0-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
https://github.com/turboderp-org/exllamav2/releases/download/v0.3.2/exllamav2-0.3.2+cu128.torch2.7.0-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11" https://github.com/turboderp-org/exllamav2/releases/download/v0.3.2/exllamav2-0.3.2+cu128.torch2.7.0-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
https://github.com/turboderp-org/exllamav2/releases/download/v0.3.2/exllamav2-0.3.2-py3-none-any.whl; platform_system == "Linux" and platform_machine != "x86_64" https://github.com/turboderp-org/exllamav2/releases/download/v0.3.2/exllamav2-0.3.2-py3-none-any.whl; platform_system == "Linux" and platform_machine != "x86_64"

4
requirements/portable/requirements.txt
View file

@ -23,5 +23,5 @@ sse-starlette==1.6.5
tiktoken tiktoken
# CUDA wheels # CUDA wheels
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+cu124-py3-none-win_amd64.whl; platform_system == "Windows" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+cu124-py3-none-win_amd64.whl; platform_system == "Windows"
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+cu124-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+cu124-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"

4
requirements/portable/requirements_amd.txt
View file

@ -23,5 +23,5 @@ sse-starlette==1.6.5
tiktoken tiktoken
# AMD wheels # AMD wheels
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+vulkan-py3-none-win_amd64.whl; platform_system == "Windows" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+vulkan-py3-none-win_amd64.whl; platform_system == "Windows"
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+rocm6.4.4-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+rocm6.4.4-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"

4
requirements/portable/requirements_amd_noavx2.txt
View file

@ -23,5 +23,5 @@ sse-starlette==1.6.5
tiktoken tiktoken
# AMD wheels # AMD wheels
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+vulkanavx-py3-none-win_amd64.whl; platform_system == "Windows" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+vulkanavx-py3-none-win_amd64.whl; platform_system == "Windows"
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+rocm6.4.4avx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+rocm6.4.4avx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"

4
requirements/portable/requirements_apple_intel.txt
View file

@ -23,5 +23,5 @@ sse-starlette==1.6.5
tiktoken tiktoken
# Mac wheels # Mac wheels
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0-py3-none-macosx_15_0_x86_64.whl; platform_system == "Darwin" and platform_release >= "24.0.0" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0-py3-none-macosx_15_0_x86_64.whl; platform_system == "Darwin" and platform_release >= "24.0.0"
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0-py3-none-macosx_14_0_x86_64.whl; platform_system == "Darwin" and platform_release >= "23.0.0" and platform_release < "24.0.0" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0-py3-none-macosx_14_0_x86_64.whl; platform_system == "Darwin" and platform_release >= "23.0.0" and platform_release < "24.0.0"

4
requirements/portable/requirements_apple_silicon.txt
View file

@ -23,5 +23,5 @@ sse-starlette==1.6.5
tiktoken tiktoken
# Mac wheels # Mac wheels
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0-py3-none-macosx_15_0_arm64.whl; platform_system == "Darwin" and platform_release >= "24.0.0" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0-py3-none-macosx_15_0_arm64.whl; platform_system == "Darwin" and platform_release >= "24.0.0"
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0-py3-none-macosx_14_0_arm64.whl; platform_system == "Darwin" and platform_release >= "23.0.0" and platform_release < "24.0.0" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0-py3-none-macosx_14_0_arm64.whl; platform_system == "Darwin" and platform_release >= "23.0.0" and platform_release < "24.0.0"

4
requirements/portable/requirements_cpu_only.txt
View file

@ -23,5 +23,5 @@ sse-starlette==1.6.5
tiktoken tiktoken
# llama.cpp (CPU only, AVX2) # llama.cpp (CPU only, AVX2)
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+cpuavx2-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+cpuavx2-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+cpuavx2-py3-none-win_amd64.whl; platform_system == "Windows" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+cpuavx2-py3-none-win_amd64.whl; platform_system == "Windows"

4
requirements/portable/requirements_cpu_only_noavx2.txt
View file

@ -23,5 +23,5 @@ sse-starlette==1.6.5
tiktoken tiktoken
# llama.cpp (CPU only, no AVX2) # llama.cpp (CPU only, no AVX2)
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+cpuavx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+cpuavx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+cpuavx-py3-none-win_amd64.whl; platform_system == "Windows" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+cpuavx-py3-none-win_amd64.whl; platform_system == "Windows"

4
requirements/portable/requirements_noavx2.txt
View file

@ -23,5 +23,5 @@ sse-starlette==1.6.5
tiktoken tiktoken
# CUDA wheels # CUDA wheels
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+cu124avx-py3-none-win_amd64.whl; platform_system == "Windows" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+cu124avx-py3-none-win_amd64.whl; platform_system == "Windows"
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+cu124avx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+cu124avx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"

4
requirements/portable/requirements_vulkan.txt
View file

@ -23,5 +23,5 @@ sse-starlette==1.6.5
tiktoken tiktoken
# Vulkan wheels # Vulkan wheels
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+vulkan-py3-none-win_amd64.whl; platform_system == "Windows" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+vulkan-py3-none-win_amd64.whl; platform_system == "Windows"
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+vulkan-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+vulkan-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"

4
requirements/portable/requirements_vulkan_noavx2.txt
View file

@ -23,5 +23,5 @@ sse-starlette==1.6.5
tiktoken tiktoken
# CUDA wheels # CUDA wheels
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+vulkanavx-py3-none-win_amd64.whl; platform_system == "Windows" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+vulkanavx-py3-none-win_amd64.whl; platform_system == "Windows"
https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.61.0/llama_cpp_binaries-0.61.0+vulkanavx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+vulkanavx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"