Qwen-3-VL-Ricoh-8B-20260227
ricoh-ai/Qwen-3-VL-Ricoh-8B-20260227
8BQwen2 Heretic artifacts1 measured conversions246 downloads
Best version85.7Pareto
sasa2000/Qwen-3-VL-Ricoh-8B-20260227-hereticby sasa2000 · Heretic v1.2.0 · GGUF, SAFETENSORS
- KL divergence
- 0.017
- Refusals
- 27/100
- Refusal rate
- 27%
- Downloads
- 246
Evidence from the model card
| KL divergence | 0.0172 | | Refusals | 27/100 |
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 | sasa2000 | 8B | 0.017 | 27/10027% | 85.7 | BestParetoGGUFSAFETENSORS | 246 |
All artifacts (2)
| Repository | Type | Formats | Metrics | Downloads | Updated |
|---|---|---|---|---|---|
| Qwen-3-VL-Ricoh-8B-20260227-heretic-Q8_0-GGUF sasa2000 | quant | GGUFQ8_0 | inheritedKL 0.017 | 195 | 2026-05-15 |
| Qwen-3-VL-Ricoh-8B-20260227-heretic sasa2000 | heretic | SAFETENSORS | published (weak match)KL 0.017 | 51 | 2026-03-31 |