nodes/1038lab-ComfyUI-MiniCPM
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Add files via upload

Browse Source
This commit is contained in:
AI Lab
2025-08-27 10:50:43 -07:00
committed by GitHub
parent 22734e7ab3
commit 91bf9fb793
4 changed files with 36 additions and 28 deletions
Download Patch File Download Diff File Expand all files Collapse all files
README.md
+13 -13
View File
@@ -4,21 +4,19 @@ A custom ComfyUI node for MiniCPM vision-language models, supporting v4, v4.5, a
**🎉 Now supports MiniCPM-V-4.5! The latest model with enhanced capabilities.**
- MiniCPM-V-4
[![MiniCPM-V-4](example_workflows/MiniCPM-V-4.jpg)](https://github.com/1038lab/ComfyUI-MiniCPM/blob/main/example_workflows/MiniCPM-V-4.json)
- MiniCPM-V-4 GGUF
[![MiniCPM-V-4-GGUF](example_workflows/MiniCPM-V-4-GGUF.jpg)](https://github.com/1038lab/ComfyUI-MiniCPM/blob/main/example_workflows/MiniCPM-V-4-GGUF.json)
- MiniCPM-V-4 Batch Images
[![MiniCPM-V-4_batchImages](example_workflows/MiniCPM-V-4_batchImages.jpg)](https://github.com/1038lab/ComfyUI-MiniCPM/blob/main/example_workflows/MiniCPM-V-4_batchImages.json)
- MiniCPM-V-4 video
[![MiniCPM-V-4_video](example_workflows/MiniCPM-V-4_video.jpg)](https://github.com/1038lab/ComfyUI-MiniCPM/blob/main/example_workflows/MiniCPM-V-4_video.json)
---
## News & Updates
- **2025/08/27**: Update ComfyUI-MIniCPM to **v1.1.0** ( [update.md](https://github.com/1038lab/ComfyUI-MiniMPC/blob/main/update.md#v110-2025-08-27) )
-Added support for **MiniCPM-V-4.5** models (Transformers)
- **2025/08/27**: Update ComfyUI-MIniCPM to **v1.1.0** ( [update.md](update.md#v110-2025-08-27) )
[![MiniCPM v4 VS v45](example_workflows/MiniCPM_v4VSv45.jpg)](example_workflows/MiniCPM_v4VSv45.json)
- Added support for **MiniCPM-V-4.5** models (Transformers)
## Features
- MiniCPM-V-4 GGUF
[![MiniCPM-V-4-GGUF](example_workflows/MiniCPM-V-4-GGUF.jpg)](example_workflows/MiniCPM-V-4-GGUF.json)
- MiniCPM-V-4 Batch Images
[![MiniCPM-V-4_batchImages](example_workflows/MiniCPM-V-4_batchImages.jpg)](example_workflows/MiniCPM-V-4_batchImages.json)
- MiniCPM-V-4 video
[![MiniCPM-V-4_video](example_workflows/MiniCPM-V-4_video.jpg)](example_workflows/MiniCPM-V-4_video.json)
- Supports **MiniCPM-V-4.5 (Transformers)** and **MiniCPM-V-4.0 (GGUF)** models
- **Latest MiniCPM-V-4.5** with enhanced capabilities via Transformers
@@ -44,12 +42,14 @@ git clone https://github.com/1038lab/comfyui-minicpm.git
Install required dependencies:
```bash
ComfyUI\python_embeded\python pip install -r ComfyUI/custom_nodes/comfyui-minicpm/requirements.txt
cd ComfyUI/custom_nodes/comfyui-minicpm
ComfyUI\python_embeded\python pip install -r requirements.txt
ComfyUI\python_embeded\python llama_cpp_install.py
```
> [!note]
> `llama-cpp-python` CUDA Installation for ComfyUI Portable
> - [llama_cpp_install.md](https://github.com/1038lab/ComfyUI-MiniCPM/blob/main/llama_cpp_install.md)
> - [llama_cpp_install.md](llama_cpp_install/llama_cpp_install.md)
---
## Supported Models
README_CN.md
+14 -13
View File
@@ -4,24 +4,21 @@
**🎉 现在支持 MiniCPM-V-4.5！最新的模型，拥有增强的能力。**
- MiniCPM-V-4
[![MiniCPM-V-4](example_workflows/MiniCPM-V-4.jpg)](https://github.com/1038lab/ComfyUI-MiniCPM/blob/main/example_workflows/MiniCPM-V-4.json)
- MiniCPM-V-4 GGUF
[![MiniCPM-V-4-GGUF](example_workflows/MiniCPM-V-4-GGUF.jpg)](https://github.com/1038lab/ComfyUI-MiniCPM/blob/main/example_workflows/MiniCPM-V-4-GGUF.json)
- MiniCPM-V-4 Batch Images
[![MiniCPM-V-4_batchImages](example_workflows/MiniCPM-V-4_batchImages.jpg)](https://github.com/1038lab/ComfyUI-MiniCPM/blob/main/example_workflows/MiniCPM-V-4_batchImages.json)
- MiniCPM-V-4 video
[![MiniCPM-V-4_video](example_workflows/MiniCPM-V-4_video.jpg)](https://github.com/1038lab/ComfyUI-MiniCPM/blob/main/example_workflows/MiniCPM-V-4_video.json)
---
## 新闻与更新
- **2025/08/27**：ComfyUI-MiniCPM 更新至 **v1.1.0**（[更新日志](https://github.com/1038lab/ComfyUI-MiniMPC/blob/main/update.md#v110-2025-08-27)）
- 新增对 **MiniCPM-V-4.5** 模型（Transformers）的支持
[![MiniCPM v4 VS v45](example_workflows/MiniCPM_v4VSv45.jpg)](example_workflows/MiniCPM_v4VSv45.json)
- 新增对 **MiniCPM-V-4.5** 模型（Transformers）的支持
## 功能特点
- MiniCPM-V-4 GGUF
[![MiniCPM-V-4-GGUF](example_workflows/MiniCPM-V-4-GGUF.jpg)](example_workflows/MiniCPM-V-4-GGUF.json)
- MiniCPM-V-4 Batch Images
[![MiniCPM-V-4_batchImages](example_workflows/MiniCPM-V-4_batchImages.jpg)](example_workflows/MiniCPM-V-4_batchImages.json)
- MiniCPM-V-4 video
[![MiniCPM-V-4_video](example_workflows/MiniCPM-V-4_video.jpg)](example_workflows/MiniCPM-V-4_video.json)
- 支持 **MiniCPM-V-4.5 (Transformers)** 和 **MiniCPM-V-4.0 (GGUF)** 模型
- **最新 MiniCPM-V-4.5** 通过 Transformers 提供增强的能力
@@ -43,13 +40,17 @@
cd ComfyUI/custom_nodes
git clone https://github.com/1038lab/comfyui-minicpm.git
```
安装所需依赖：
```bash
pip install -r ComfyUI/custom_nodes/comfyui-minicpm/requirements.txt
cd ComfyUI/custom_nodes/comfyui-minicpm
ComfyUI\python_embeded\python pip install -r requirements.txt
ComfyUI\python_embeded\python llama_cpp_install.py
```
> [!note]
> `llama-cpp-python` CUDA Installation for ComfyUI Portable
> - [llama_cpp_install_zh.md](llama_cpp_install/llama_cpp_install_zh.md)
---
## 支持的模型
pyproject.toml
+3 -2
View File
@@ -1,7 +1,7 @@
[project]
name = "ComfyUI-MiniCPM"
description = "A ComfyUI custom node for MiniCPM vision-language models, enabling high-quality image captioning and analysis."
version = "1.0.0"
description = "A custom ComfyUI node for MiniCPM vision-language models, supporting v4, v4.5, and v4 GGUF formats, enabling high-quality image captioning and visual analysis."
version = "1.1.0"
license = {file = "LICENSE"}
dependencies = [
@@ -10,6 +10,7 @@ dependencies = [
"torchvision>=0.15.0",
"Pillow>=9.0.0",
"huggingface-hub>=0.16.0",
"hf_xet>=0.16.0",
"numpy>=1.21.0",
"accelerate>=0.20.0"
]
update.md
+6
View File
@@ -2,6 +2,7 @@
## V1.1.0 (2025-08-27)
### New Features
[![MiniCPM v4 VS v45](example_workflows/MiniCPM_v4VSv45.jpg)](example_workflows/MiniCPM_v4VSv45.json)
- **MiniCPM-V-4.5 Model Support**: Added comprehensive support for the latest MiniCPM-V-4.5 models
- New Transformers model: `MiniCPM-V-4.5` (openbmb/MiniCPM-V-4_5) - Latest full precision version
- New quantized model: `MiniCPM-V-4.5-int4` (openbmb/MiniCPM-V-4_5-int4) - Memory-efficient 4-bit version
@@ -21,19 +22,24 @@
## V1.0.0 (2025-08-25)
### Initial Release
- **MiniCPM-V-4 Transformers Support**: Full integration with Hugging Face transformers models
[![MiniCPM-V-4](example_workflows/MiniCPM-V-4.jpg)](example_workflows/MiniCPM-V-4.json)
- MiniCPM-V-4 full precision model
- MiniCPM-V-4-int4 quantized model
- **MiniCPM-V-4 GGUF Support**: Comprehensive GGUF model support via llama-cpp-python
[![MiniCPM-V-4-GGUF](example_workflows/MiniCPM-V-4-GGUF.jpg)](example_workflows/MiniCPM-V-4-GGUF.json)
- Multiple quantization levels (Q4_0 to Q8_0)
- Optimized memory usage and performance
- **Dual Node Architecture**:
- Basic nodes for simple usage
- Advanced nodes with full parameter control
- **Image & Video Processing**: Support for both image and video input processing
[![MiniCPM-V-4_video](example_workflows/MiniCPM-V-4_video.jpg)](example_workflows/MiniCPM-V-4_video.json)
- **Memory Management**: Three modes - Keep in Memory, Clear After Run, Global Cache
- **Caption Types**: 14 different preset prompt types for various use cases
- **Legacy Compatibility**: Backward compatible node for existing workflows
[![MiniCPM-V-4_batchImages](example_workflows/MiniCPM-V-4_batchImages.jpg)](example_workflows/MiniCPM-V-4_batchImages.json)
### Core Features
- **Multi-Modal Processing**: Image and video content understanding
- **Flexible Parameters**: Customizable temperature, top-p, top-k, repetition penalty