Diffusion Single File
GGUF
comfyui
conversational
How to use from
Pi
Start the llama.cpp server
# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf on-nurida-mov/My-MiniMax-H3:Q4_0
Configure the model in Pi
# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:9931/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "on-nurida-mov/My-MiniMax-H3:Q4_0"
        }
      ]
    }
  }
}
Run Pi
# Start Pi in your project directory:
pi
Quick Links

MiniMax H3

Repackaged model files for ComfyUI.

Original model repository: https://huggingface.co/MiniMaxAI/MiniMax-H3

Place the files in the following folders:

πŸ“‚ ComfyUI/
β”œβ”€β”€ πŸ“‚ models/
β”‚   β”œβ”€β”€ πŸ“‚ diffusion_models/
β”‚   β”‚   β”œβ”€β”€ minimax_h3_fl2va_bf16.safetensors
β”‚   β”‚   β”œβ”€β”€ minimax_h3_fl2va_int8_convrot.safetensors
β”‚   β”‚   β”œβ”€β”€ minimax_h3_fl2va_pruned_bf16.safetensors
β”‚   β”‚   β”œβ”€β”€ minimax_h3_fl2va_pruned_int8_convrot.safetensors
β”‚   β”‚   β”œβ”€β”€ minimax_h3_fl2va_pruned_fp8_scaled.safetensors
β”‚   β”‚   β”œβ”€β”€ minimax_h3_ref2va_bf16.safetensors
β”‚   β”‚   β”œβ”€β”€ minimax_h3_ref2va_int8_convrot.safetensors
β”‚   β”‚   β”œβ”€β”€ minimax_h3_ref2va_pruned_bf16.safetensors
β”‚   β”‚   β”œβ”€β”€ minimax_h3_ref2va_pruned_int8_convrot.safetensors
β”‚   β”‚   └── minimax_h3_ref2va_pruned_fp8_scaled.safetensors
β”‚   β”œβ”€β”€ πŸ“‚ text_encoders/
β”‚   β”‚   β”œβ”€β”€ qwen3vl_32b_minimax_h3_bf16.safetensors
β”‚   β”‚   β”œβ”€β”€ qwen3vl_32b_minimax_h3_int8_convrot.safetensors
β”‚   β”‚   └── qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors
β”‚   β”œβ”€β”€ πŸ“‚ vae/
β”‚   β”‚   β”œβ”€β”€ minimax_h3_audio_vae_fp32.safetensors
β”‚   β”‚   └── minimax_h3_video_vae_fp16.safetensors

Workflows

Downloads last month
980
GGUF
Model size
12B params
Architecture
gemma4
Hardware compatibility
Log In to add your hardware

4-bit

Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support

Model tree for on-nurida-mov/My-MiniMax-H3

Quantized
(96)
this model