File size: 1,648 Bytes
10cebb2 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 | ---
license: other
license_name: modified-mit
license_link: https://huggingface.co/moonshotai/Kimi-K2.7-Code/blob/main/LICENSE
base_model: moonshotai/Kimi-K2.7-Code
tags:
- mlx
- vision
- kimi
- exo
---
# Kimi-K2.7-Code-vision
Vision-only weights (MoonViT tower + multimodal projector) extracted from
[moonshotai/Kimi-K2.7-Code](https://huggingface.co/moonshotai/Kimi-K2.7-Code)
for use with MLX-based inference stacks such as [exo](https://github.com/exo-explore/exo),
in the same format as [exolabs/Kimi-K2.6-vision](https://huggingface.co/exolabs/Kimi-K2.6-vision).
## Contents
- `kimi_k27_vision.safetensors` — all 335 `vision_tower.*` and `mm_projector.*`
tensors from the official repo (shards 63–64), original bfloat16, unmodified.
- `config.json` — vision config copied from the official `config.json`
(verified byte-identical to Kimi-K2.6's vision config: 27-layer MoonViT,
hidden 1152, patch 14, `sd2_tpool` merger, projector to 7168).
- `extract_vision_weights.py` — the script used to produce this repo,
for reproducibility.
## Usage with exo
Add a model card for `moonshotai/Kimi-K2.7-Code` with:
```toml
capabilities = ["text", "thinking", "thinking_toggle", "vision"]
[vision]
image_token_id = 163605
model_type = "kimi_vl"
weights_repo = "aidiffuser/Kimi-K2.7-Code-vision"
processor_repo = "moonshotai/Kimi-K2.7-Code"
```
Tested working: distributed (2× Mac Studio M3 Ultra, tensor parallelism) with
the official INT4 text weights, image understanding confirmed.
## License
Same Modified MIT license as the source model; these are a subset of the
original weights, unmodified. All credit to Moonshot AI.
|