Qwen2.5-VL-32B-Instruct
Qwen/Qwen2.5-VL-32B-Instruct
32BQwen 2.53 Heretic artifacts1 measured conversions12.5K downloads
Best version78.4Pareto
coder3101/Qwen2.5-VL-32B-Instruct-hereticby coder3101 · Heretic v1.1.0 · GGUF, SAFETENSORS
- KL divergence
- 0.412
- Refusals
- 14/100
- Refusal rate
- 14%
- Downloads
- 12.5K
Evidence from the model card
| KL divergence | 0.4118 | | Refusals | 14/100 |
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 | coder3101 | 32B | 0.412 | 14/10014% | 78.4 | BestParetoGGUFSAFETENSORS | 12.5K |
All artifacts (3)
| Repository | Type | Formats | Metrics | Downloads | Updated |
|---|---|---|---|---|---|
| Qwen2.5-VL-32B-Instruct-heretic-i1-GGUF mradermacher | quant | GGUFIQ1_M | inheritedKL 0.412 | 8.1K | 2026-01-18 |
| Qwen2.5-VL-32B-Instruct-heretic-GGUF mradermacher | quant | GGUFIQ4_XS | inheritedKL 0.412 | 4.2K | 2026-10-08 |
| Qwen2.5-VL-32B-Instruct-heretic coder3101 | heretic | SAFETENSORS | published (weak match)KL 0.412 | 208 | 2026-01-17 |