FuseChat-Qwen-2.5-7B-Instruct
FuseAI/FuseChat-Qwen-2.5-7B-Instruct
7BQwen4 Heretic artifacts1 measured conversions25.2K downloads
Best version95.2Pareto
ChiKoi7/FuseChat-Qwen-2.5-7B-Instruct-Hereticby ChiKoi7 · Heretic v1.0.1 · GGUF, SAFETENSORS
- KL divergence
- 0.070
- Refusals
- 3/100
- Refusal rate
- 3%
- Downloads
- 25.2K
Evidence from the model card
| Refusals | 3/100 | | KL divergence | 0.07 |
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 | ChiKoi7 | 7B | 0.070 | 3/1003% | 95.2 | BestParetoGGUFSAFETENSORS | 25.2K |
All artifacts (4)
| Repository | Type | Formats | Metrics | Downloads | Updated |
|---|---|---|---|---|---|
| FuseChat-Qwen-2.5-7B-Instruct-Heretic-i1-GGUF mradermacher | quant | GGUFIQ1_M | inheritedKL 0.070 | 16.7K | 2025-12-11 |
| FuseChat-Qwen-2.5-7B-Instruct-Heretic-GGUF ChiKoi7 | heretic | GGUF | published (weak match)KL 0.070 | 5.4K | 2025-12-10 |
| FuseChat-Qwen-2.5-7B-Instruct-Heretic-GGUF mradermacher | quant | GGUFIQ4_XS | inheritedKL 0.070 | 2.9K | 2025-12-11 |
| FuseChat-Qwen-2.5-7B-Instruct-Heretic ChiKoi7 | heretic | SAFETENSORS | published (weak match)KL 0.070 | 159 | 2025-12-10 |