SWE-bench Verified Mini (HAL)
HAL's cost-aware agent leaderboard for the SWE-bench Verified Mini software engineering subset.
33rows
accuracyprimary metric
2026-05-27sampled
Metadata
Metrics
Accuracy, Cost (USD) (lower is better), Runs
| Rank | Subject | Accuracy | Model Match | Provenance | Sampled |
|---|---|---|---|---|---|
| 1 | SWE-Agent / Claude Sonnet 4.5 High (September 2025) | 72 | — | Verified | 2026-05-27 |
| 2 | SWE-Agent / Claude Sonnet 4.5 (September 2025) | 68 | — | Verified | 2026-05-27 |
| 3 | SWE-Agent / Claude Opus 4.1 (August 2025) | 61 | — | Verified | 2026-05-27 |
| 4 | SWE-Agent / o4-mini Low (April 2025) | 54 | — | Verified | 2026-05-27 |
| 5 | SWE-Agent / Claude-3.7 Sonnet High (February 2025) | 54 | — | Verified | 2026-05-27 |
| 6 | SWE-Agent / Claude Opus 4.1 High (August 2025) | 54 | — | Verified | 2026-05-27 |
| 7 | SWE-Agent / o4-mini High (April 2025) | 50 | — | Verified | 2026-05-27 |
| 8 | SWE-Agent / Claude-3.7 Sonnet (February 2025) | 50 | — | Verified | 2026-05-27 |
| 9 | SWE-Agent / Claude Opus 4 (May 2025) | 50 | — | Verified | 2026-05-27 |
| 10 | SWE-Agent / GPT-5 Medium (August 2025) | 46 | — | Verified | 2026-05-27 |
| 11 | HAL Generalist Agent / Claude Opus 4.1 High (August 2025) | 46 | — | Verified | 2026-05-27 |
| 12 | SWE-Agent / o3 Medium (April 2025) | 46 | — | Verified | 2026-05-27 |
| 13 | HAL Generalist Agent / Claude Haiku 4.5 High (October 2025) | 44 | — | Verified | 2026-05-27 |
| 14 | SWE-Agent / GPT-4.1 (April 2025) | 44 | — | Verified | 2026-05-27 |
| 15 | HAL Generalist Agent / Claude Opus 4.1 (August 2025) | 42 | — | Verified | 2026-05-27 |
| 16 | HAL Generalist Agent / Claude Sonnet 4.5 High (September 2025) | 40 | — | Verified | 2026-05-27 |
| 17 | HAL Generalist Agent / Claude Sonnet 4.5 (September 2025) | 34 | — | Verified | 2026-05-27 |
| 18 | HAL Generalist Agent / Claude Opus 4 (May 2025) | 34 | — | Verified | 2026-05-27 |
| 19 | HAL Generalist Agent / Claude Opus 4 High (May 2025) | 30 | — | Verified | 2026-05-27 |
| 20 | HAL Generalist Agent / Claude-3.7 Sonnet (February 2025) | 26 | — | Verified | 2026-05-27 |
| 21 | SWE-Agent / Gemini 2.0 Flash (February 2025) | 24 | — | Verified | 2026-05-27 |
| 22 | SWE-Agent / DeepSeek V3 (March 2025) | 24 | — | Verified | 2026-05-27 |
| 23 | HAL Generalist Agent / Claude-3.7 Sonnet High (February 2025) | 24 | — | Verified | 2026-05-27 |
| 24 | HAL Generalist Agent / Claude Haiku 4.5 (October 2025) | 24 | — | Verified | 2026-05-27 |
| 25 | HAL Generalist Agent / GPT-5 Medium (August 2025) | 12 | — | Verified | 2026-05-27 |
| 26 | HAL Generalist Agent / DeepSeek V3 (March 2025) | 10 | — | Verified | 2026-05-27 |
| 27 | HAL Generalist Agent / o4-mini Low (April 2025) | 6 | — | Verified | 2026-05-27 |
| 28 | HAL Generalist Agent / DeepSeek R1 (January 2025) | 6 | — | Verified | 2026-05-27 |
| 29 | HAL Generalist Agent / Gemini 2.0 Flash (February 2025) | 2 | — | Verified | 2026-05-27 |
| 30 | HAL Generalist Agent / o4-mini High (April 2025) | 2 | — | Verified | 2026-05-27 |
| 31 | HAL Generalist Agent / GPT-4.1 (April 2025) | 2 | — | Verified | 2026-05-27 |
| 32 | SWE-Agent / DeepSeek R1 (January 2025) | 0 | — | Verified | 2026-05-27 |
| 33 | HAL Generalist Agent / o3 Medium (April 2025) | 0 | — | Verified | 2026-05-27 |
No matching rows.