agents benchmark
SWE-fficiency Leaderboard
SWE-fficiency is an open-source benchmark and workflow that evaluates language models on optimizing the runtime efficiency of real-world software engineering tasks, measuring how well agents can improve code performance autonomously.
Updated Aug 17, 2026
Models1
Model coverage1
MetricScore
EvidenceB
SWE-fficiency Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelMI | Score34.8% | Percentile100.0% | Participants1 | EvidenceC | Evaluated |
SWE-fficiency Highlights
The leading models and scores on this benchmark.
The Top AI Models for SWE-fficiency
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis swe-fficiency AI model leaderboard uses descending score in the benchmark's original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.
- 01MIMiniMax M3MiniMax
- Score
- 34.8%
- Price
- $0.30 input / $1.2 output per 1M tokens
- Speed
- Up to 214 tok/s via Together
Strengths
- Ranks #1 of 1 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures SWE-fficiency, not total model capability
Selection summary
Best AI Models for SWE-fficiency
MiniMax M3 currently leads SWE-fficiency with 34.8%. It is the top model on this specific benchmark, while the best LLM for the broader task should also be checked against other benchmarks, price and runtime.
Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.
What is SWE-fficiency?
What SWE-fficiency measures and how its scores work.
SWE-fficiency is an open-source benchmark and workflow that evaluates language models on optimizing the runtime efficiency of real-world software engineering tasks, measuring how well agents can improve code performance autonomously.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- SWE-fficiency
- Modality
- text
- Primary category
- agents
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- swe-fficiency|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about SWE-fficiency.
Which model scores highest on SWE-fficiency?
MiniMax M3 is currently ranked first with 34.8%.
What does SWE-fficiency measure?
SWE-fficiency is an open-source benchmark and workflow that evaluates language models on optimizing the runtime efficiency of real-world software engineering tasks, measuring how well agents can improve code performance autonomously.
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.