wezzel98765
/

Qwen3.6-35B-A3B-oQ4e-fp16-mtp

4-bit precision

Model card Files Files and versions

wezzel98765 commited on Jul 10

Commit

1071116

·

verified ·

1 Parent(s): 121054f

Update README.md

Files changed (1) hide show

README.md +9 -0

README.md CHANGED Viewed

@@ -1,3 +1,12 @@
 ---
 license: apache-2.0
 ---

 ---
 license: apache-2.0
+library_name: mlx
+base_model:
+- Qwen/Qwen3.6-35B-A3B
 ---
+Text and vision retained<br>
+Quantised using oMLX v0.5.0.rc1 OQ Enhanced quantization (oQe) iMatrix<br>
+MTP Heads retained <br>
+FP16 is fastest on M1/M2 , but this can work on all MLX inferencing systems <br>
+This model is using the LATEST FROGGERIC chat template upgrade (Fixed jinja chat templates for Qwen 3.5 & 3.6 (v21)) <br>