Is Hopper 12B Trained SOTA in Decision models?#120 in Decision modelsRanked #120 in Decision models.
How it scores here
Hopper 12B Trained in Decision models
Primary benchmarks
- JevBench: 68.1 (CI 67.6–68.6) on Oct 8, 2026 (#37 of 134).
On the category's weighted score it is 27.3 points behind #1 karpo-decision-31B and 0.0 behind the next model up, Eikos-27B.
Nearby on the board
| # | Model | Category score |
|---|---|---|
| 117 | garas (Gemma 4 31B) | 73.3 |
| 118 | Shisa DE-2.1 (repeat) | 73.0 |
| 119 | Eikos-27B | 72.8 |
| 120 | Hopper 12B Trained | 72.7 |
| 121 | reflex Qwen3.8-27B-FP8 (pmscr) | 72.5 |
| 122 | RSI-Jev v6.1 VL 27B (effort auto) | 72.2 |
| 123 | SPX-CD-Omni (SurdAI, google/gemma-4-12B-it + LoRA r32, one forward pass, self-hosted) | 72.0 |
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 |