| --- |
| 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. |
|
|
