Is JevK5 v0.3 (4B) SOTA in Decision models?#158 in Decision modelsRanked #158 in Decision models.
How it scores here
JevK5 v0.3 (4B) in Decision models
Primary benchmarks
- JevBench: 64.2 (CI 63.7–64.7) on Oct 8, 2026 (#49 of 134).
On the category's weighted score it is 35.6 points behind #1 karpo-decision-31B and 0.2 behind the next model up, StartLux-Decision v1.1 4B.
Nearby on the board
| # | Model | Category score |
|---|---|---|
| 155 | decider-4b v2 (Mapika) | 65.2 |
| 156 | GS1 9B | 64.9 |
| 157 | StartLux-Decision v1.1 4B | 64.6 |
| 158 | JevK5 v0.3 (4B) | 64.4 |
| 159 | Decider chat · Qwen3.6-27B | 64.3 |
| 160 | Perplexity Decider Flash v1.2 (4B) | 64.1 |
| 161 | Jev-Omni (akhilaaa3, Gemma-4-12B merged) | 64.0 |
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 |