LFM2-VL-3B
LiquidAI/LFM2-VL-3B
3BLFM5 Heretic artifacts1 measured conversions14.9K downloads
Best version97.0Pareto
ZuzeTt/LFM2-VL-3B-heretic-Imatrix-GGUFby ZuzeTt · Heretic v1.0.1 · GGUF, SAFETENSORS
- KL divergence
- 0.020
- Refusals
- 4/100
- Refusal rate
- 4%
- Downloads
- 14.9K
Evidence from the model card
| KL divergence | 0.02 | | Refusals | 4/100 |
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 | ZuzeTt | 3B | 0.020 | 4/1004% | 97.0 | BestParetoGGUFSAFETENSORS | 14.9K |
All artifacts (5)
| Repository | Type | Formats | Metrics | Downloads | Updated |
|---|---|---|---|---|---|
| LFM2-VL-3B-heretic-Imatrix-GGUF ZuzeTt | heretic | GGUFF16 | published (weak match)KL 0.020 | 9.5K | 2026-02-23 |
| LFM2-VL-3B-heretic-GGUF MuXodious | quant | GGUFBF16 | published (weak match)KL 0.020 | 2.3K | 2026-01-05 |
| LFM2-VL-3B-heretic-imatrix-GGUF MuXodious | quant | GGUFBF16 | published (weak match)KL 0.020 | 1.6K | 2026-01-06 |
| LFM2-VL-3B-heretic-Q8_0-GGUF DiffusionFanatic1 | quant | GGUFQ8_0 | inheritedKL 0.020 | 1.3K | 2025-11-20 |
| LFM2-VL-3B-heretic pszemraj | heretic | SAFETENSORS | published (weak match)KL 0.020 | 209 | 2025-12-29 |