Kimi-VL-A3B-Thinking-2506-8bit
mlx-community/Kimi-VL-A3B-Thinking-2506-8bit
Vision + text · transformers
8-bit Open on Hugging Face
Parameters
—
Weights (est.)
—
Mac memory (est.)
—
Size is estimated from the parameter count and quantisation. Memory allows a fifth extra for context plus 2 GB for the system; long contexts need more.
Run it
With mlx-vlm on an Apple silicon Mac:
pip install mlx-vlmpython -m mlx_vlm.generate --model mlx-community/Kimi-VL-A3B-Thinking-2506-8bit --image photo.jpg --prompt "Describe this image"
Details
- Base model
- moonshotai/Kimi-VL-A3B-Instruct (quantized)
- Vendor
- Moonshot · Kimi
- Task
- Vision + text
- Quantisation
- 8-bit
- Licence
- mit
- Downloads
- 199 in the last 30 days · 3.2K all time · 2 likes
- Published
- 2025-08-16 · updated 2026-01-27 (9mo ago)
- Tags
- kimi_vl, feature-extraction, custom_code
Other MLX builds of moonshotai/Kimi-VL-A3B-Instruct
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Kimi-VL-A3B-Thinking-2506-bf16 | bf16 | ≈ 32.8 GB | 48 GB+ | 148 |
| Kimi-VL-A3B-Thinking-2506-8bit (this one) | 8-bit | — | — | 199 |
| Kimi-VL-A3B-Thinking-2506-6bit | 6-bit | — | — | 198 |
| Kimi-VL-A3B-Thinking-2506-5bit | 5-bit | — | — | 150 |