# Kimi-VL-A3B-Thinking
A powerful vision-language model with extended thinking capabilities from Moonshot AI.
## Key Features
- **Vision & Reasoning** - Understands images and uses chain-of-thought reasoning
- **128K Context** - Massive 131,072 token context window
- **MoE Architecture** - 64 experts + 2 shared experts for efficient inference
- **DeepSeek2 Base** - Built on proven DeepSeek2 architecture with MLA attention
## Quick Start
```bash
ollama run richardyoung/kimi-vl-a3b-thinking
```
## Available Formats
| Format | Size | Description |
|--------|------|-------------|
| Q4_K_M (default) | 9.8 GB | Best balance of quality and speed |
| F16 | 30 GB | Full precision |
## Technical Requirements
- **Minimum:** 16GB RAM
- **Recommended:** 32GB RAM or Apple Silicon Mac with 24GB+ unified memory
## Links
- **Original Model:** moonshotai/Kimi-VL-A3B-Thinking-2506
- **GGUF Quantizations:** richardyoung/Kimi-VL-A3B-Thinking-GGUF
## Credits
- **Original Model:** Moonshot AI
- **Quantization:** Richard Young (deepneuro.ai/richard)
## License
MIT License