mlx-community
/

FuseO1-DeepSeekR1-Qwen2.5-Coder-32B-Preview-Q8

8-bit precision

Model card Files Files and versions

bobig commited on Feb 20

Commit

27de11c

·

verified ·

1 Parent(s): 94da5f0

Update README.md

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -5,9 +5,9 @@ tags:
 - mlx
 ---
-13.7 TPS
-27.1 TPS with Speculative decoding in LMstudio.
 Draft model: [DeepScaleR-1.5B-Preview-Q8](https://huggingface.co/mlx-community/DeepScaleR-1.5B-Preview-Q8)

 - mlx
 ---
+13 TPS
+27 TPS with Speculative decoding in LMstudio.
 Draft model: [DeepScaleR-1.5B-Preview-Q8](https://huggingface.co/mlx-community/DeepScaleR-1.5B-Preview-Q8)