chore: prepare v2.3.23 release

__init__.py

@@ -3,7 +3,7 @@
     "name": "ComfyUI-ThinkingLLM",
     "description": "A multimodal ComfyUI AI node with Qwen3.5, Qwen3-VL, Qwen2.5-VL, Qwen3, and Gemma 4 integrations. Features live thinking in the terminal to see what the LLM is doing in real time.",
     "author": "goodguy1963",
-    "version": "2.3.22",
+    "version": "2.3.23",
     "url": "https://github.com/goodguy1963/ComfyUI-ThinkingLLM",
     "category": "image"
 }

custom_nodes.json

@@ -9,5 +9,5 @@
   "description": "A multimodal ComfyUI AI node with Qwen3.5, Qwen3-VL, Qwen2.5-VL, Qwen3, and Gemma 4 integrations. Features live thinking in the terminal to see what the LLM is doing in real time.",
   "category": "image",
   "tags": ["vision", "language", "multimodal", "thinkingllm", "qwen", "gguf", "flash_attention", "performance", "wan2.2", "video", "i2v", "t2v", "storyboard", "caching", "fork", "gpl-3.0"],
-  "version": "2.3.22"
+  "version": "2.3.23"
 }

pyproject.toml

@@ -1,7 +1,7 @@
 [project]
 name = "ComfyUI-ThinkingLLM"
 description = "A multimodal ComfyUI AI node with Qwen3.5, Qwen3-VL, Qwen2.5-VL, Qwen3, and Gemma 4 integrations. Features live thinking in the terminal to see what the LLM is doing in real time."
-version = "2.3.22"
+version = "2.3.23"
 license = {file = "LICENSE"}
 dependencies = ["transformers>=5.2.0", "torch", "huggingface-hub", "hf_xet", "psutil", "numpy", "Pillow", "opencv-python", "bitsandbytes", "accelerate", "sentencepiece", "tiktoken", "faster-whisper", "llama-cpp-python"]
 

update.md

@@ -1,5 +1,19 @@
 # ComfyUI-QwenVL Update Log
 
+## Version 2.3.23 (2026/08/08) - MiniMax H3 Reference Prompt Guard
+
+### 2.3.23 Changes
+
+- Required an explicit target-video prompt and reference-role assignment for MiniMax H3 Reference-to-Video generation.
+- Applied the same validation to Hugging Face, ComfyUI-native, and GGUF vision nodes.
+- Rejected empty ComfyUI-native model generations before they can be cached or saved as blank output.
+- Documented the correct Text-to-Video versus Reference-to-Video selection and recommended direct-generation settings.
+
+### 2.3.23 Validation
+
+- Added regression coverage for missing Reference-to-Video target prompts and empty native model output.
+- All 114 isolated regression tests pass.
+
 ## Version 2.3.22 (2026/08/06) - Community Model Catalog Fix
 
 ### 2.3.22 Changes