Qwen3.5-0.8B-Claude-4.6-Opus-Reasoning-Distilled
Jackrong/Qwen3.5-0.8B-Claude-4.6-Opus-Reasoning-Distilled
0.8BQwen 3.52 Heretic artifacts1 measured conversions5K downloads
Best version94.9Pareto
dalatexcoder/Qwen3.5-0.8B-Claude-4.6-Opus-Reasoning-Distilled-hereticby dalatexcoder · Heretic v1.2.0 · LORA, SAFETENSORS
- KL divergence
- 0.045
- Refusals
- 6/100
- Refusal rate
- 6%
- Downloads
- 659
Evidence from the model card
| KL divergence | 0.0446 | | Refusals | 6/100 |
Metrics published by the author. Why this ranks first.
All measured versions
Higher score is better. One row per ablation attempt; artifacts of the same attempt are grouped under it.
| # | Author | Params | KL | Refusals | Score | Evidence | Downloads |
|---|---|---|---|---|---|---|---|
| 1 | dalatexcoder | 0.8B | 0.045 | 6/1006% | 94.9 | BestParetoLORASAFETENSORS | 659 |
All artifacts (2)
| Repository | Type | Formats | Metrics | Downloads | Updated |
|---|---|---|---|---|---|
| Jackrong-Qwen3.5-0.8B-Claude-4.6-Opus-Reasoning-Distiled-heretic-GGUF mradermacher | quant | GGUFIQ4_XS | none | 4.3K | 2026-03-26 |
| Qwen3.5-0.8B-Claude-4.6-Opus-Reasoning-Distilled-heretic dalatexcoder | heretic | LORASAFETENSORS | published (weak match)KL 0.045 | 659 | 2026-04-28 |
1 artifact inherit metrics from their parent conversion or publish none.