VERIFIEDNov 222026
Is Amber 4B state of the art?#272Best current rank: #272 in Decision models, as of Nov 22, 2026.
As of Nov 22, 2026, Amber 4B is ranked #272 in Decision models (since Nov 22, 2026; sources Decision Index Full; methodology v1.0). Ranked #272 in Decision models.
At a glance
Facts and scores
Amber 4B is an open-weights text model from Unknown; 4.2B parameters and no public price.
How it scores
- On Decision Index ECE it scores 5.7%, #178 of 420 (top 43%); 4.6 pts behind KnowLine 9B Gen2; frontier-infra Jebadiah 27B v2.1 (t1e) is 0.1 pts above and level with Nace.AI Drex Max v1 (27B) below.
- On Decision Index Brier it scores 38.1%, #206 of 419 (top 50%); 17.7 pts behind lg-decision-4B; GS1 0.8B is 0.1 pts above and level with PKU-KCL-Meteor-0.8B below.
- On Decision Index Full it scores 39.5 (CI 39.0–40.0), #258 of 420 (top 62%); 34.4 behind karpo-decision-31B; JPT-4B is 0.1 above and ClassOne Qwen 3.5 9B (Qwen3.5-9B backbone with ClassOne decision heads, schema-constrained single forward pass, self-hosted) is 0.4 below.
History
First result on isitsota: Nov 21, 2026. Best overall rank so far: #272 in Decision models, first held Nov 22, 2026.
Calibration
Reliability
Index bins for Amber 4B. Confidence on the horizontal axis, accuracy on the vertical axis.
Median 127.6 ms, same GPU.
Receipts
Evidence
| Source | Score | Date | Type | Receipt |
|---|---|---|---|---|
| Decision Index Full Amber 4B | 39.5 39.0–40.0 | 2026-11-21 | INDEP. | snap_decision-index_2026-11-21_release-v4|decision-index-0.4.1 |
Reign history
Where it sits in the succession
Crown intervals for this model are still empty.
Honest about the gaps
Where it is not the crown
| Category | This model | Rank | Leader |
|---|---|---|---|
| Decision models | #272 in this category | #272 | karpo-decision-31B |
Links
Related pages
Is Amber 4B SOTA?
Best current rank: #272 in Decision models, as of Nov 22, 2026.
What is Amber 4B best at?
As of Nov 22, 2026, Amber 4B is ranked #272 in Decision models (since Nov 22, 2026; sources Decision Index Full).