Other · model
SWE-2 Flash benchmark scores
As of 2026-11-05, SWE-2 Flash (Other) has measured scores on 7 benchmarks and estimated scores on 20 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-SciCode | 55.1% | estimated ± 3.5 pp, medium confidence |
| SciCode | 54.8% | estimated ± 7.6 pp, low confidence |
| Terminal-Bench 2.1 | 89.2% | measured |
| AA Coding Index | 77.1% | estimated ± 2.4 pp, low confidence |
| SWE-bench Verified | 87.7% | estimated ± 0.8 pp, medium confidence |
| SWE-bench Pro | 66.0% | estimated ± 7.4 pp, low confidence |
| AA Terminal-Bench 2.1 | 85.9% | estimated ± 6.8 pp, medium confidence |
| CursorBench 3.2 | 62.5% | estimated ± 1.6 pp, medium confidence |
| Terminal-Bench 3.0 | 25.4% | estimated ± 9.1 pp, low confidence |
| FrontierCode 1.1 Main | 45.5% | measured |
| FrontierSWE v2 | 40.0% | estimated ± 8.9 pp, medium confidence |
| Vals LiveCodeBench | 86.9% | estimated ± 1.9 pp, high confidence |
| Vals SWE-bench | 85.6% | estimated ± 2.8 pp, high confidence |
| Terminal-Bench 2.1 (Vals AI) | 75.8% | estimated ± 5.4 pp, low confidence |
| AA Terminal-Bench 4.0 | 19.9% | estimated ± 5.5 pp, medium confidence |
| Bug Hunt Bench | 30.6% | estimated ± 8.4 pp, low confidence |
| CursorBench 4.0 | 41.9% | estimated ± 2.7 pp, low confidence |
| DeepSWE | 60.1% | measured |
| FrontierCode 2.0 | 24.0% | measured |
| PostTrainBench v1.1 | 28.8% | estimated ± 4.5 pp, high confidence |
| Terminal-Bench 4.0 | 20.0% | measured |
| Vibe Code Bench | 74.4% | estimated ± 5.6 pp, low confidence |
| ExploitBench | 34.8% | measured |
| NL2Repo | 56.1% | estimated ± 6.2 pp, medium confidence |
| ScarfBench | 69.9% | estimated ± 3.8 pp, medium confidence |
| SEC-Bench Pro | 55.9% | estimated ± 3.9 pp, medium confidence |
| SWE-Atlas Codebase QnA | 50.0% | measured |