Qwen3-VL-2B-Instruct
Qwen/Qwen3-VL-2B-Instruct
2BQwen 36 Heretic artifacts4 measured conversions9.3K downloads
Best version95.0Pareto
saidutta69/Qwen3-VL-2B-Instruct-hereticby saidutta69 · Heretic v1.4.0 · GGUF, SAFETENSORS
- KL divergence
- 0.053
- Refusals
- 5/100
- Refusal rate
- 5%
- Downloads
- 2K
Evidence from the model card
| KL divergence | 0.0528 | | Refusals | 5/100 |
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 | saidutta69 | 2B | 0.053 | 5/1005% | 95.0 | BestParetoGGUFSAFETENSORS | 2K |
| 2 | ChiKoi7 | 2B | 0.055 | 6/1006% | 94.4 | SAFETENSORS | 79 |
| 3 | sh0ck0r | 2B | 0.067 | 11/10011% | 91.4 | SAFETENSORS | 81 |
| 4 | coder3101 | 2B | 0.450 | 2/1002% | 83.5 | ParetoSAFETENSORS | 1.1K |
All artifacts (6)
| Repository | Type | Formats | Metrics | Downloads | Updated |
|---|---|---|---|---|---|
| Qwen-3-VL-2B-Instruct-heretic-i1-GGUF mradermacher | quant | GGUFIQ1_M | none | 3.1K | 2026-08-06 |
| Qwen-3-VL-2B-Instruct-heretic-GGUF mradermacher | quant | GGUFIQ4_XS | none | 2.8K | 2026-10-07 |
| Qwen3-VL-2B-Instruct-heretic saidutta69 | quant | GGUFSAFETENSORSF16 | published (weak match)KL 0.053 | 2K | 2026-10-05 |
| Qwen3-VL-2B-Instruct-heretic coder3101 | heretic | SAFETENSORS | published (weak match)KL 0.450 | 1.1K | 2026-01-17 |
| Qwen3-VL-2B-Instruct-heretic sh0ck0r | heretic | SAFETENSORS | published (weak match)KL 0.067 | 81 | 2026-03-07 |
| Qwen3-VL-2B-Instruct-Heretic ChiKoi7 | heretic | SAFETENSORS | published (weak match)KL 0.055 | 79 | 2026-04-06 |
2 artifacts inherit metrics from their parent conversion or publish none.