agents benchmark
SWE Atlas - Codebase QnA
SWE Atlas - Codebase QnA evaluates a model's ability to answer questions about real codebases, measuring repository-level comprehension and the ability to reason about code structure, behavior, and intent across an entire project.
Updated Aug 11, 2026
Models1
Model coverage1
MetricScore
EvidenceB
SWE Atlas - Codebase QnA Ranking
Higher score ranks better on this benchmark.
| 01 | MI | 37.9% | 100.0% | 1 | C |
SWE Atlas - Codebase QnA Highlights
The leading models and scores on this benchmark.
What is SWE Atlas - Codebase QnA?
What SWE Atlas - Codebase QnA measures and how its scores work.
SWE Atlas - Codebase QnA evaluates a model's ability to answer questions about real codebases, measuring repository-level comprehension and the ability to reason about code structure, behavior, and intent across an entire project.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- SWE Atlas - Codebase QnA
- Modality
- text
- Primary category
- agents
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- swe-atlas-codebase-qna|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about SWE Atlas - Codebase QnA.
Which model scores highest on SWE Atlas - Codebase QnA?
MiniMax M3 is currently ranked first with 37.9%.
What does SWE Atlas - Codebase QnA measure?
SWE Atlas - Codebase QnA evaluates a model's ability to answer questions about real codebases, measuring repository-level comprehension and the ability to reason about code structure, behavior, and intent across an entire project.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
1 model results are currently shown.
Does this benchmark affect the overall score?
No. This benchmark is shown for reference but does not contribute to the overall score.