Downloads · 30 days
53
2% of all-time downloads
Kimi-K2.6_dq3-mlx
cs2764/Kimi-K2.6_dq3-mlx
Kimi-K2.6_dq3-mlx is a feature extraction model from cs2764. Use it when you need embeddings to search or compare text. It is set up for mlx.
This model is a DQ3 quantized version of the original model moonshotai/Kimi-K2.6. It was quantized locally using the mlxlm library.
All-time downloads
3K
Public
Parameters
1T
469 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors469 GB · 100%
Parameter types
How the weights are stored.
U321T · 100%
Model card
From the Hugging Face model README
Kimi-K2.6_dq3
This model is a DQ3 quantized version of the original model moonshotai/Kimi-K2.6.
It was quantized locally using the mlx_lm library.
Quantization Methodology (DQ3)
This model was quantized using the dynamic DQ3 (3-bit / 4-bit / 8-bit mixed) approach, inspired by the methodology described in the mlx-community/Kimi-K2.5-mlx-DQ3_K_M-q8 repository.
The weights are mixed based on MLX layers:
- Expert layers (switch_mlp / mlp) are quantized to 3-bit.
- The first 5 layers are kept at higher quality (5-bit).
- Every 5th layer is medium quality (4-bit).
- All other layers (e.g. attention, normalization) remain at 8-bit to serve as the "8-bit brain".
- Task
- Feature Extraction
- Library
- mlx
- Type
- kimi_k25
- Created
- Apr 24, 2026
- Updated
- Apr 24, 2026