Is Amber 0.4B v2 SOTA in Decision models?#341 in Decision modelsRanked #341 in Decision models.
How it scores here
Amber 0.4B v2 in Decision models
Primary benchmarks
- Decision Index Full: 27.8 (CI 27.3–28.3) on Nov 21, 2026 (#325 of 420).
On the category's weighted score it is 75.9 points behind #1 karpo-decision-31B and 0.2 behind the next model up, smalljev semantic-v9.
Nearby on the board
| # | Model | Category score |
|---|---|---|
| 338 | openvons | 24.6 |
| 339 | ClassOne Qwen3.5 4B | 24.3 |
| 340 | smalljev semantic-v9 | 24.2 |
| 341 | Amber 0.4B v2 | 24.1 |
| 342 | Valen-Decision 1.5 0.8B | 23.8 |
| 343 | jul fast (usejul/jul-decision-minicpm5-2b: MiniCPM5-2B fine-tune with a pointer head, option probabilities from one pass, self-hosted via `jul serve`) | 23.5 |
| 344 | ClassOne Gemma 4 E4B | 23.5 |
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 |