code benchmark
SWT-Bench
Software Test Benchmark evaluating LLM ability to write tests for software repositories
Updated Aug 7, 2026
Published models1
Registry coverage1
MetricScore
EvidenceB
SWT-Bench leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | MI | 69.3% | 100.0% | 1 | C |
SWT-Bench highlights
The top published results on this benchmark's own scale.
What is SWT-Bench?
Definition and scoring fields from the benchmark registry.
Software Test Benchmark evaluating LLM ability to write tests for software repositories
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- SWT-Bench
- Modality
- text
- Primary category
- code
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- swt-bench|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about SWT-Bench.
Which model scores highest on SWT-Bench?
MiniMax M2.1 is currently ranked first with 69.3%.
What does SWT-Bench measure?
Software Test Benchmark evaluating LLM ability to write tests for software repositories
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
1 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.