reasoning benchmark
SWE-bench Verified (Multiple Attempts) Leaderboard
SWE-bench Verified (Multiple Attempts) is a human-validated benchmark of 500 test samples from the original SWE-bench dataset for evaluating AI systems on real GitHub issues in Python repositories.
Updated Sep 5, 2026
Models1
Model coverage1
MetricScore
EvidenceB
SWE-bench Verified (Multiple Attempts) Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelMA | Score71.60% | Percentile100.00% | Participants1 | EvidenceC | Evaluated |
SWE-bench Verified (Multiple Attempts) Highlights
The leading models and scores on this benchmark.
The Top AI Models for SWE-bench Verified (Multiple Attempts)
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is SWE-bench Verified (Multiple Attempts)?
What SWE-bench Verified (Multiple Attempts) measures and how its scores work.
SWE-bench Verified is a human-validated subset of 500 test samples from the original SWE-bench dataset in which models edit codebases to resolve issue descriptions.
It measures AI systems' ability to automatically resolve real GitHub issues in Python repositories, including changes across multiple functions, classes, and files.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- SWE-bench Verified (Multiple Attempts)
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- overall
LLM Stats. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about SWE-bench Verified (Multiple Attempts).
Which model scores highest on SWE-bench Verified (Multiple Attempts)?
Kimi K2 Instruct is currently ranked first with 71.60%.
What are the top three models on SWE-bench Verified (Multiple Attempts)?
The current leaders are Kimi K2 Instruct (71.60%).
Which SWE-bench Verified (Multiple Attempts) model has the lowest official input price?
No matched official input price is currently available.
Which models are fastest among SWE-bench Verified (Multiple Attempts) results?
No matched runtime record is currently available.
Does the highest score result prove overall model quality?
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.