Other · model
Ember-1 benchmark scores
As of 2026-11-21, Ember-1 (Other) has measured scores on 4 benchmarks and estimated scores on 27 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-SciCode | 53.8% | estimated ± 3.8 pp, high confidence |
| SciCode | 53.0% | estimated ± 3.6 pp, medium confidence |
| Terminal-Bench 2.1 | 82.0% | measured |
| AA Coding Index | 77.8% | estimated ± 2.5 pp, high confidence |
| SWE-bench Verified | 92.2% | measured |
| SWE-bench Pro | 72.0% | estimated ± 4.5 pp, high confidence |
| Vals LiveCodeBench | 90.8% | estimated ± 10.0 pp, low confidence |
| AA Terminal-Bench 2.1 | 81.7% | estimated ± 2.6 pp, high confidence |
| CursorBench 3.2 | 68.2% | estimated ± 6.4 pp, low confidence |
| Terminal-Bench 3.0 | 21.4% | estimated ± 3.2 pp, high confidence |
| FrontierCode 1.1 Main | 52.5% | estimated ± 3.5 pp, low confidence |
| Terminal-Bench 4.1 | 21.8% | measured |
| Terminal-Bench Hard | 58.5% | estimated ± 2.8 pp, medium confidence |
| Vals SWE-bench | 92.5% | estimated ± 6.6 pp, medium confidence |
| Terminal-Bench 2.1 (Vals AI) | 69.1% | estimated ± 7.3 pp, medium confidence |
| Bug Hunt Bench | 37.7% | estimated ± 10.4 pp, low confidence |
| DeepSWE | 75.2% | measured |
| FrontierCode 2.0 | 32.3% | estimated ± 10.2 pp, low confidence |
| PostTrainBench v1.1 | 43.6% | estimated ± 4.5 pp, high confidence |
| Terminal-Bench 4.0 | 28.1% | estimated ± 4.5 pp, high confidence |
| Terminal-Bench-Science 0.1 | 0.0% | estimated ± 10.7 pp, low confidence |
| VulcanBench Frontier v4 | 90.6% | estimated ± 5.0 pp, high confidence |
| AA Terminal-Bench 4.0 | 21.4% | estimated ± 6.2 pp, medium confidence |
| Vibe Code Bench | 66.3% | estimated ± 8.4 pp, low confidence |
| NL2Repo | 63.0% | estimated ± 6.0 pp, medium confidence |
| PostTrainBench v1 | 35.4% | estimated ± 5.5 pp, low confidence |
| React Native Evals | 85.1% | estimated ± 2.7 pp, low confidence |
| SWE-Rebench | 61.8% | estimated ± 3.6 pp, low confidence |
| ScarfBench | 69.4% | estimated ± 3.8 pp, medium confidence |
| LiveCodeBench | 86.2% | estimated ± 14.6 pp, low confidence |
| SWE-Atlas Codebase QnA | 61.7% | estimated ± 4.8 pp, high confidence |