Is Raw Phi-4 mini direct logits SOTA in Decision models?#386 in Decision modelsRanked #386 in Decision models.
How it scores here
Raw Phi-4 mini direct logits in Decision models
Primary benchmarks
- JevBench: 35.3 (CI 34.8–35.8) on Oct 8, 2026 (#115 of 134).
On the category's weighted score it is 85.6 points behind #1 karpo-decision-31B and 0.1 behind the next model up, vLLM-SR Decision 2.0 Eos 0.8B.
Nearby on the board
| # | Model | Category score |
|---|---|---|
| 383 | ClassOne Qwen3.5 2B | 15.1 |
| 384 | JEV Ettin 150M | 14.8 |
| 385 | vLLM-SR Decision 2.0 Eos 0.8B | 14.5 |
| 386 | Raw Phi-4 mini direct logits | 14.4 |
| 387 | Crivo-150M v2 | 14.2 |
| 388 | Bekko System One v0 400M | 13.9 |
| 389 | Raw Qwen3 4B Instruct 2507 direct logits | 13.6 |
Standings
Decision models
| # | Model | State |
|---|---|---|
| 1 | karpo-decision-31B | Co-SOTA |
| 1 | Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout) | Co-SOTA |
| 1 | GLiDE 27B | Co-SOTA |
| 1 | kudhab (Gemma 4 31B) | Co-SOTA |
| 2 | Quyet-1.0-Large (Chinh Nguyen, Gemma-4-31B decoder, option-letter logits) | Near-SOTA |
| 3 | REAX 1.3 | Near-SOTA |
| 4 | Logits Sigma-1 Duo | Near-SOTA |
| 5 | KnowLine 31B Gen2 | Near-SOTA |
| 6 | Strands Decider v2 31B | Near-SOTA |
| 7 | Copernicus-31B | Near-SOTA |
| 8 | decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted) | Near-SOTA |
| 9 | Copernicus-57B | Near-SOTA |