Image: Add placeholder file for user_data/image_models

js/switch_tabs.js

@@ -36,3 +36,17 @@ function switch_to_character() {
   document.getElementById("character-tab-button").click();
   scrollToTop();
 }
+
+function switch_to_image_ai_generate() {
+  const container = document.querySelector("#image-ai-tab");
+  const buttons = container.getElementsByTagName("button");
+
+  for (let i = 0; i < buttons.length; i++) {
+    if (buttons[i].textContent.trim() === "Generate") {
+      buttons[i].click();
+      break;
+    }
+  }
+
+  scrollToTop();
+}

modules/chat.py

@@ -3,7 +3,6 @@ import copy
 import functools
 import html
 import json
-import os
 import pprint
 import re
 import shutil
@@ -26,6 +25,7 @@ from modules.html_generator import (
     convert_to_markdown,
     make_thumbnail
 )
+from modules.image_utils import open_image_safely
 from modules.logging_colors import logger
 from modules.text_generation import (
     generate_reply,
@@ -1516,20 +1516,6 @@ def load_instruction_template_memoized(template):
     return load_instruction_template(template)
 
 
-def open_image_safely(path):
-    if path is None or not isinstance(path, str) or not Path(path).exists():
-        return None
-
-    if os.path.islink(path):
-        return None
-
-    try:
-        return Image.open(path)
-    except Exception as e:
-        logger.error(f"Failed to open image file: {path}. Reason: {e}")
-        return None
-
-
 def upload_character(file, img_path, tavern=False):
     img = open_image_safely(img_path)
     decoded_file = file if isinstance(file, str) else file.decode('utf-8')

modules/image_models.py

@@ -98,7 +98,7 @@ def load_image_model(model_name, dtype='bfloat16', attn_backend='sdpa', cpu_offl
     Args:
         model_name: Name of the model directory
         dtype: 'bfloat16' or 'float16'
-        attn_backend: 'sdpa', 'flash_attention_2', or 'flash_attention_3'
+        attn_backend: 'sdpa' or 'flash_attention_2'
         cpu_offload: Enable CPU offloading for low VRAM
         compile_model: Compile the model for faster inference (slow first run)
         quant_method: 'none', 'bnb-8bit', 'bnb-4bit', or torchao options (int8wo, fp4, float8wo)
@@ -145,8 +145,6 @@ def load_image_model(model_name, dtype='bfloat16', attn_backend='sdpa', cpu_offl
         if hasattr(pipe, 'transformer') and hasattr(pipe.transformer, 'set_attention_backend'):
             if attn_backend == 'flash_attention_2':
                 pipe.transformer.set_attention_backend("flash")
-            elif attn_backend == 'flash_attention_3':
-                pipe.transformer.set_attention_backend("_flash_3")
             # sdpa is the default, no action needed
 
         if compile_model:

modules/image_utils.py

@@ -1,9 +1,7 @@
-"""
-Shared image processing utilities for multimodal support.
-Used by both ExLlamaV3 and llama.cpp implementations.
-"""
 import base64
 import io
+import os
+from pathlib import Path
 from typing import Any, List, Tuple
 
 from PIL import Image
@@ -11,6 +9,20 @@ from PIL import Image
 from modules.logging_colors import logger
 
 
+def open_image_safely(path):
+    if path is None or not isinstance(path, str) or not Path(path).exists():
+        return None
+
+    if os.path.islink(path):
+        return None
+
+    try:
+        return Image.open(path)
+    except Exception as e:
+        logger.error(f"Failed to open image file: {path}. Reason: {e}")
+        return None
+
+
 def convert_pil_to_base64(image: Image.Image) -> str:
     """Converts a PIL Image to a base64 encoded string."""
     buffered = io.BytesIO()

modules/shared.py

@@ -56,7 +56,7 @@ group = parser.add_argument_group('Image model')
 group.add_argument('--image-model', type=str, help='Name of the image model to select on startup (overrides saved setting).')
 group.add_argument('--image-model-dir', type=str, default='user_data/image_models', help='Path to directory with all the image models.')
 group.add_argument('--image-dtype', type=str, default=None, choices=['bfloat16', 'float16'], help='Data type for image model.')
-group.add_argument('--image-attn-backend', type=str, default=None, choices=['sdpa', 'flash_attention_2', 'flash_attention_3'], help='Attention backend for image model.')
+group.add_argument('--image-attn-backend', type=str, default=None, choices=['sdpa', 'flash_attention_2'], help='Attention backend for image model.')
 group.add_argument('--image-cpu-offload', action='store_true', help='Enable CPU offloading for image model.')
 group.add_argument('--image-compile', action='store_true', help='Compile the image model for faster inference.')
 group.add_argument('--image-quant', type=str, default=None,

modules/ui_image_generation.py

@@ -7,7 +7,6 @@ from pathlib import Path
 
 import gradio as gr
 import numpy as np
-from PIL import Image
 from PIL.PngImagePlugin import PngInfo
 
 from modules import shared, ui, utils
@@ -16,6 +15,7 @@ from modules.image_models import (
     load_image_model,
     unload_image_model
 )
+from modules.image_utils import open_image_safely
 from modules.logging_colors import logger
 from modules.text_generation import stop_everything_event
 from modules.utils import gradio
@@ -29,7 +29,7 @@ ASPECT_RATIOS = {
 }
 
 STEP = 16
-IMAGES_PER_PAGE = 64
+IMAGES_PER_PAGE = 32
 
 # Settings keys to save in PNG metadata (Generate tab only)
 METADATA_SETTINGS_KEYS = [
@@ -159,7 +159,7 @@ def save_generated_images(images, state, actual_seed):
 def read_image_metadata(image_path):
     """Read generation metadata from PNG file."""
     try:
-        with Image.open(image_path) as img:
+        with open_image_safely(image_path) as img:
             if hasattr(img, 'text') and 'image_gen_settings' in img.text:
                 return json.loads(img.text['image_gen_settings'])
     except Exception as e:
@@ -172,7 +172,7 @@ def format_metadata_for_display(metadata):
     if not metadata:
         return "No generation settings found in this image."
 
-    lines = ["**Generation Settings**", ""]
+    lines = []
 
     # Display in a nice order
     display_order = [
@@ -418,9 +418,9 @@ def create_ui():
                         # Pagination controls
                         with gr.Row():
                             shared.gradio['image_refresh_history'] = gr.Button("🔄 Refresh", elem_classes="refresh-button")
-                            shared.gradio['image_prev_page'] = gr.Button("◀ Prev", elem_classes="refresh-button")
+                            shared.gradio['image_prev_page'] = gr.Button("◀ Prev Page", elem_classes="refresh-button")
                             shared.gradio['image_page_info'] = gr.Markdown(value=get_initial_page_info, elem_id="image-page-info")
-                            shared.gradio['image_next_page'] = gr.Button("Next ▶", elem_classes="refresh-button")
+                            shared.gradio['image_next_page'] = gr.Button("Next Page ▶", elem_classes="refresh-button")
                             shared.gradio['image_page_input'] = gr.Number(value=1, label="Page", precision=0, minimum=1, scale=0, min_width=80)
                             shared.gradio['image_go_to_page'] = gr.Button("Go", elem_classes="refresh-button", scale=0, min_width=50)
 
@@ -441,7 +441,7 @@ def create_ui():
                         )
 
                     with gr.Column(scale=1):
-                        gr.Markdown("### Selected Image")
+                        gr.Markdown("### Generation Settings")
                         shared.gradio['image_settings_display'] = gr.Markdown("Select an image to view its settings")
                         shared.gradio['image_send_to_generate'] = gr.Button("Send to Generate", variant="primary")
                         shared.gradio['image_gallery_status'] = gr.Markdown("")
@@ -485,7 +485,7 @@ def create_ui():
                                     info='bfloat16 recommended for modern GPUs'
                                 )
                                 shared.gradio['image_attn_backend'] = gr.Dropdown(
-                                    choices=['sdpa', 'flash_attention_2', 'flash_attention_3'],
+                                    choices=['sdpa', 'flash_attention_2'],
                                     value=shared.settings['image_attn_backend'],
                                     label='Attention Backend',
                                     info='SDPA is default. Flash Attention requires compatible GPU.'
@@ -649,6 +649,7 @@ def create_event_handlers():
             'image_cfg_scale',
             'image_gallery_status'
         ),
+        js=f'() => {{{ui.switch_tabs_js}; switch_to_image_ai_generate()}}',
         show_progress=False
     )
 
@@ -676,7 +677,8 @@ def generate(state):
 
         if not model_name or model_name == 'None':
             logger.error("No image model selected. Go to the Model tab and select a model.")
-            return []
+            yield []
+            return
 
         if shared.image_model is None:
             result = load_image_model(
@@ -689,7 +691,8 @@ def generate(state):
             )
             if result is None:
                 logger.error(f"Failed to load model `{model_name}`.")
-                return []
+                yield []
+                return
 
             shared.image_model_name = model_name
 
@@ -759,6 +762,7 @@ def generate(state):
             generator.manual_seed(int(seed + i))
             batch_results = shared.image_model(**gen_kwargs).images
             all_images.extend(batch_results)
+            yield all_images
 
         t1 = time.time()
         save_generated_images(all_images, state, seed)
@@ -767,12 +771,12 @@ def generate(state):
         total_steps = state["image_steps"] * int(state['image_batch_count'])
         logger.info(f'Generated {total_images} {"image" if total_images == 1 else "images"} in {(t1 - t0):.2f} seconds ({total_steps / (t1 - t0):.2f} steps/s, seed {seed})')
 
-        return all_images
+        yield all_images
 
     except Exception as e:
         logger.error(f"Image generation failed: {e}")
         traceback.print_exc()
-        return []
+        yield []
 
 
 def load_image_model_wrapper(model_name, dtype, attn_backend, cpu_offload, compile_model, quant_method):

requirements/full/requirements.txt

@@ -44,8 +44,8 @@ sse-starlette==1.6.5
 tiktoken
 
 # CUDA wheels
-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/v0.62.0/llama_cpp_binaries-0.62.0+cu124-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+cu124-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+cu124-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
 https://github.com/turboderp-org/exllamav3/releases/download/v0.0.16/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/v0.0.16/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"

requirements/full/requirements_amd.txt

@@ -42,7 +42,7 @@ sse-starlette==1.6.5
 tiktoken
 
 # AMD wheels
-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/v0.62.0/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/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+vulkan-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+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-py3-none-any.whl; platform_system != "Darwin" and platform_machine != "x86_64"

requirements/full/requirements_amd_noavx2.txt

@@ -42,7 +42,7 @@ sse-starlette==1.6.5
 tiktoken
 
 # AMD wheels
-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/v0.62.0/llama_cpp_binaries-0.62.0+vulkanavx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+vulkanavx-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+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-py3-none-any.whl; platform_system != "Darwin" and platform_machine != "x86_64"

requirements/full/requirements_apple_intel.txt

@@ -42,5 +42,5 @@ sse-starlette==1.6.5
 tiktoken
 
 # Mac wheels
-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/v0.62.0/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"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0-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/v0.64.0/llama_cpp_binaries-0.64.0-py3-none-macosx_14_0_x86_64.whl; platform_system == "Darwin" and platform_release >= "23.0.0" and platform_release < "24.0.0"

requirements/full/requirements_apple_silicon.txt

@@ -42,5 +42,5 @@ sse-starlette==1.6.5
 tiktoken
 
 # Mac wheels
-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/v0.62.0/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"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0-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/v0.64.0/llama_cpp_binaries-0.64.0-py3-none-macosx_14_0_arm64.whl; platform_system == "Darwin" and platform_release >= "23.0.0" and platform_release < "24.0.0"

requirements/full/requirements_cpu_only.txt

@@ -42,5 +42,5 @@ sse-starlette==1.6.5
 tiktoken
 
 # llama.cpp (CPU only, AVX2)
-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/v0.62.0/llama_cpp_binaries-0.62.0+cpuavx2-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+cpuavx2-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+cpuavx2-py3-none-win_amd64.whl; platform_system == "Windows"

requirements/full/requirements_cpu_only_noavx2.txt

@@ -42,5 +42,5 @@ sse-starlette==1.6.5
 tiktoken
 
 # llama.cpp (CPU only, no AVX2)
-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/v0.62.0/llama_cpp_binaries-0.62.0+cpuavx-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+cpuavx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+cpuavx-py3-none-win_amd64.whl; platform_system == "Windows"

requirements/full/requirements_noavx2.txt

@@ -44,8 +44,8 @@ sse-starlette==1.6.5
 tiktoken
 
 # CUDA wheels
-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/v0.62.0/llama_cpp_binaries-0.62.0+cu124avx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+cu124avx-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+cu124avx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
 https://github.com/turboderp-org/exllamav3/releases/download/v0.0.16/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/v0.0.16/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"

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/v0.62.0/llama_cpp_binaries-0.62.0+cu124-py3-none-win_amd64.whl; platform_system == "Windows"
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+cu124-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+cu124-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+cu124-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"

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/v0.62.0/llama_cpp_binaries-0.62.0+vulkan-py3-none-win_amd64.whl; platform_system == "Windows"
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/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/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+vulkan-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+rocm6.4.4-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"

requirements/portable/requirements_amd_noavx2.txt

@@ -23,5 +23,5 @@ sse-starlette==1.6.5
 tiktoken
 
 # AMD wheels
-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/v0.62.0/llama_cpp_binaries-0.62.0+rocm6.4.4avx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+vulkanavx-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+rocm6.4.4avx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"

requirements/portable/requirements_apple_intel.txt

@@ -23,5 +23,5 @@ sse-starlette==1.6.5
 tiktoken
 
 # Mac wheels
-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/v0.62.0/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"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0-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/v0.64.0/llama_cpp_binaries-0.64.0-py3-none-macosx_14_0_x86_64.whl; platform_system == "Darwin" and platform_release >= "23.0.0" and platform_release < "24.0.0"

requirements/portable/requirements_apple_silicon.txt

@@ -23,5 +23,5 @@ sse-starlette==1.6.5
 tiktoken
 
 # Mac wheels
-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/v0.62.0/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"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0-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/v0.64.0/llama_cpp_binaries-0.64.0-py3-none-macosx_14_0_arm64.whl; platform_system == "Darwin" and platform_release >= "23.0.0" and platform_release < "24.0.0"

requirements/portable/requirements_cpu_only.txt

@@ -23,5 +23,5 @@ sse-starlette==1.6.5
 tiktoken
 
 # llama.cpp (CPU only, AVX2)
-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/v0.62.0/llama_cpp_binaries-0.62.0+cpuavx2-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+cpuavx2-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+cpuavx2-py3-none-win_amd64.whl; platform_system == "Windows"

requirements/portable/requirements_cpu_only_noavx2.txt

@@ -23,5 +23,5 @@ sse-starlette==1.6.5
 tiktoken
 
 # llama.cpp (CPU only, no AVX2)
-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/v0.62.0/llama_cpp_binaries-0.62.0+cpuavx-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+cpuavx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+cpuavx-py3-none-win_amd64.whl; platform_system == "Windows"

requirements/portable/requirements_noavx2.txt

@@ -23,5 +23,5 @@ sse-starlette==1.6.5
 tiktoken
 
 # CUDA wheels
-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/v0.62.0/llama_cpp_binaries-0.62.0+cu124avx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+cu124avx-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+cu124avx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"

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/v0.62.0/llama_cpp_binaries-0.62.0+vulkan-py3-none-win_amd64.whl; platform_system == "Windows"
-https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.62.0/llama_cpp_binaries-0.62.0+vulkan-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+vulkan-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+vulkan-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"

requirements/portable/requirements_vulkan_noavx2.txt

@@ -23,5 +23,5 @@ sse-starlette==1.6.5
 tiktoken
 
 # CUDA wheels
-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/v0.62.0/llama_cpp_binaries-0.62.0+vulkanavx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+vulkanavx-py3-none-win_amd64.whl; platform_system == "Windows"
+https://github.com/oobabooga/llama-cpp-binaries/releases/download/v0.64.0/llama_cpp_binaries-0.64.0+vulkanavx-py3-none-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64"