Qwen2.5-Coder-7B-Instruct
Qwen/Qwen2.5-Coder-7B-Instruct
7BQwen 2.57 Heretic artifacts2 measured conversions42.3K downloads
Best version98.3Pareto
trohrbaugh/Qwen2.5-Coder-7B-Instruct-hereticby trohrbaugh · Heretic v1.2.0 · GGUF, SAFETENSORS
- KL divergence
- 0.014
- Refusals
- 2/100
- Refusal rate
- 2%
- Downloads
- 18.6K
Evidence from the model card
| KL divergence | 0.0136 | | Refusals | 2/100 |
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 | trohrbaugh | 7B | 0.014 | 2/1002% | 98.3 | BestParetoGGUFSAFETENSORS | 18.6K |
| 2 | saidutta69 | 7B | 0.020 | 3/1003% | 97.5 | GGUFSAFETENSORS | 8.4K |
All artifacts (7)
| Repository | Type | Formats | Metrics | Downloads | Updated |
|---|---|---|---|---|---|
| Qwen2.5-Coder-7B-Instruct-heretic-i1-GGUF mradermacher | quant | GGUFIQ1_M | inheritedKL 0.014 | 12.6K | 2026-04-10 |
| HERETICODER-2.5-7B-IT-i1-GGUF mradermacher | quant | GGUFIQ1_M | none | 9.7K | 2025-12-05 |
| Qwen2.5-Coder-7B-Instruct-heretic saidutta69 | heretic | GGUFSAFETENSORSF16 | published (weak match)KL 0.020 | 8.4K | 2026-09-13 |
| Qwen2.5-Coder-7B-Instruct-heretic-GGUF mradermacher | quant | GGUFIQ4_XS | inheritedKL 0.014 | 5.2K | 2026-04-10 |
| qwen-coder-jailbreak agustindxm | heretic | GGUFF16 | noneKL 0.034 | 3.6K | 2026-08-18 |
| HERETICODER-2.5-7B-IT-GGUF mradermacher | quant | GGUFIQ4_XS | none | 2.1K | 2025-12-01 |
| Qwen2.5-Coder-7B-Instruct-heretic trohrbaugh | heretic | SAFETENSORS | published (weak match)KL 0.014 | 734 | 2026-04-14 |
3 artifacts inherit metrics from their parent conversion or publish none.