Mythos-nano
squ11z1/Mythos-nano
3.1BMythos3 Heretic artifacts1 measured conversions4.8K downloads
- KL divergence
- 0.061
- Refusals
- 13/100
- Refusal rate
- 13%
- Downloads
- 401
Evidence from the model card
| Refusals after | 13 / 100 | | KL divergence | 0.061 |
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 | richardyoung | 3.1B | 0.061 | 13/10013% | 90.6 | BestParetoMLXSAFETENSORS | 401 |
All artifacts (3)
| Repository | Type | Formats | Metrics | Downloads | Updated |
|---|---|---|---|---|---|
| Mythos-nano-heretic-i1-GGUF mradermacher | quant | GGUFIQ1_M | none | 3.3K | 2026-06-27 |
| Mythos-nano-heretic-GGUF mradermacher | quant | GGUFIQ4_XS | none | 1.1K | 2026-06-27 |
| mythos-nano-abliterated-mlx richardyoung | quant | MLXSAFETENSORS | published (weak match)KL 0.061 | 401 | 2026-06-26 |
2 artifacts inherit metrics from their parent conversion or publish none.