tcclaviger
/

Qwen3.6-40B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-Thinking-FP8-MTP

Model card Files Files and versions

tcclaviger commited on May 5

Commit

67bfcd1

·

verified ·

1 Parent(s): 4ed961b

Update README.md

Files changed (1) hide show

README.md +1 -0

README.md CHANGED Viewed

@@ -40,6 +40,7 @@ pipeline_tag: image-text-to-text
 base_model:
 - DavidAU/Qwen3.6-40B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-Thinking
 ---
 <I>The Qwen 3.5 version (also 40B) got 181 likes+ This version uses the new Qwen 3.6 27B arch (which exceeds even Qwen's own 398B model).</I>

 base_model:
 - DavidAU/Qwen3.6-40B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-Thinking
 ---
+My only change was selective FP8 quantization and insertion of MTP, otherwise exactly as David delivered.
 <I>The Qwen 3.5 version (also 40B) got 181 likes+ This version uses the new Qwen 3.6 27B arch (which exceeds even Qwen's own 398B model).</I>