LFM2.5-VL-1.6B-heretic-imatrix-GGUF
FadedRedStar/LFM2.5-VL-1.6B-heretic-imatrix-GGUF
quantGGUFIQ4_NL
Published metrics
- KL divergence
- 0.011
- Refusals
- 8/100
- Original refusals
- 95/100
- Heretic version
- v1.3.0
Published on the model card
Matched snippet
| KL divergence | 0.0114 | | Refusals | 8/100 |
Provenance
- Canonical model
- LiquidAI/LFM2.5-VL-1.6B
- Declared parent(s)
- coder3101/LFM2.5-VL-1.6B-heretic
- Classification
- quantized artifact of a Heretic model
- Created / updated
- 2026-07-03 / 2026-07-10
- Downloads
- 2K all time · 158 last 30 days
- License
- other
Refusal rate 8%. Metrics are self-reported by the publisher; the directory does not re-run evaluations.