DeepSeek-R1-Distill-Qwen-32B
deepseek-ai/DeepSeek-R1-Distill-Qwen-32B
32BQwen4 Heretic artifacts1 measured conversions19.7K downloads
Best version97.1Pareto
blackbook-lm/DeepSeek-R1-Distill-Qwen-32B-hereticby blackbook-lm · Heretic v1.2.0 · GGUF, SAFETENSORS
- KL divergence
- 0.029
- Refusals
- 3/100
- Refusal rate
- 3%
- Downloads
- 19.7K
Evidence from the model card
| KL divergence | 0.0291 | | Refusals | 3/100 |
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 | blackbook-lm | 32B | 0.029 | 3/1003% | 97.1 | BestParetoGGUFSAFETENSORS | 19.7K |
All artifacts (4)
| Repository | Type | Formats | Metrics | Downloads | Updated |
|---|---|---|---|---|---|
| DeepSeek-R1-Distill-Qwen-32B-heretic-i1-GGUF mradermacher | quant | GGUFIQ1_M | inheritedKL 0.029 | 10.2K | 2026-03-03 |
| DeepSeek-R1-Distill-Qwen-32B-heretic-GGUF mradermacher | quant | GGUFIQ4_XS | inheritedKL 0.029 | 6.3K | 2026-03-02 |
| DeepSeek-R1-Distill-Qwen-32B-heretic-i1-GGUF fairy322 | quant | GGUFIQ1_M | inheritedKL 0.029 | 2.8K | 2026-06-15 |
| DeepSeek-R1-Distill-Qwen-32B-heretic blackbook-lm | heretic | SAFETENSORS | published (weak match)KL 0.029 | 427 | 2026-03-01 |