Skip to content
AI Atlas
BenchmarkActivecategory · coding

SWE-bench Verified

swebench.com

resolve real GitHub issues (500 human-validated instances)

quality57

Updated 8 h ago · first seen 11 Sept 2026

bench_01M293SPEGA1XMKH0SMTFTCK2R

Metric
resolved · %
Direction
Higher is better
Results
180 · 10 filtered
Leader
Claude Opus 4.5 79.2%

Score history · Multiple 22 rows

0%20%40%60%80%Jan 25Apr 25Jul 25Oct 25
  • Multiple
  • 73.8%date=2025-11-03 · board=Verified · system=Salesforce AI Research SAGE · model_tag=claude-sonnet-4.53 Nov 2025
  • 73%date=2025-10-21 · board=Verified · system=Salesforce AI Research SAGE · model_tag=claude-sonnet-4.521 Oct 2025
  • 74.6%date=2025-09-15 · board=Verified · system=JoyCode · model_tag=claude-4-sonnet15 Sept 2025
  • 76.8%date=2025-09-02 · board=Verified · system=Atlassian Rovo Dev · model_tag=claude-sonnet-4-202505142 Sept 2025
  • 75.6%date=2025-09-01 · board=Verified · system=Warp · model_tag=gpt-51 Sept 2025
  • 76.4%date=2025-08-19 · board=Verified · system=ACoder · model_tag=claude-4-sonnet19 Aug 2025
  • 71.2%date=2025-07-15 · board=Verified · system=Qodo Command · model_tag=claude-sonnet-4-2025051415 Jul 2025
  • 71%date=2025-06-23 · board=Verified · system=Warp · model_tag=claude-sonnet-4-2025051423 Jun 2025
  • 75.2%date=2025-06-12 · board=Verified · system=TRAE · model_tag=claude-4-sonnet-2025052212 Jun 2025
  • 74.4%date=2025-06-03 · board=Verified · system=Refact.ai Agent · model_tag=claude-4-sonnet3 Jun 2025
  • 70.2%date=2025-05-19 · board=Verified · system=devlo · model_tag=claude-3-7-sonnet-2025021919 May 2025
  • 68.2%date=2025-05-16 · board=Verified · system=Nemotron-CORTEXA · model_tag=NV-EmbedCode16 May 2025

Back to the full leaderboard

Leaderboard 10 current results · config contains “OpenHands”

Select models with +, then open Compare.

Leaderboard
#ModelScoreConfigEvaluatedSourceActions
1#1MultipleAnthropic73.8%date=2025-11-03 · board=Verified · system=Salesforce AI Research SAGE · model_tag=claude-sonnet-4.53 Nov 2025swebench.comT2 History
2#2gpt-5OpenAI71.8%date=2025-08-07 · board=Verified · system=OpenHands · model_tag=openai/gpt-5-2025-08-077 Aug 2025swebench.comT2 History
3#3Claude 4 SonnetAnthropic70.4%date=2025-05-24 · board=Verified · system=OpenHands · model_tag=claude-4-sonnet-2025051424 May 2025swebench.comT2 History
4#4qwen3-coder-480b-a35b-instructAlibaba Group69.6%date=2025-08-05 · board=Verified · system=OpenHands · model_tag=https://huggingface.co/Qwen/Qwen3-Coder-480B-A35B-Instruct5 Aug 2025swebench.comT2 History
5#5Undisclosed65.8%date=2025-04-15 · board=Verified · system=OpenHands · submission=20250415_openhands15 Apr 2025swebench.comT2 History
6#6Kimi K2 0711Moonshot AI65.4%date=2025-07-16 · board=Verified · system=OpenHands · model_tag=moonshot/kimi-k2-0711-preview16 Jul 2025swebench.comT2 History
7#74x Scaled60.8%date=2025-02-03 · board=Verified · system=OpenHands · submission=20250203_openhands_4x_scaled3 Feb 2025swebench.comT2 History
8#8CodeAct v2.1 (claude-3-5-sonnet-20241022)Anthropic53%date=2024-10-29 · board=Verified · system=OpenHands · submission=20241029_OpenHands-CodeAct-2.1-sonnet-2024102229 Oct 2024swebench.comT2 History
9#9Qwen3 Coder 30B A3B InstructQwen51.6%date=2025-08-05 · board=Verified · system=OpenHands · model_tag=https://huggingface.co/Qwen/Qwen3-Coder-30B-A3B-Instruct5 Aug 2025swebench.comT2 History
10#10Devstral Small 1.0Mistral AI46.8%date=2025-05-20 · board=Verified · system=OpenHands · model_tag=mistralai/Devstral-Small-250520 May 2025swebench.comT2 History

10 results

Scores are reported as published, with their evaluation configuration (harness, prompting, judge). The bar is relative to the best score on this page. Results with different configs are not directly comparable — see methodology.

The config filter matches a value inside each result's configuration (server-side, `config=` on the API). Chips are the values shared by several rows on the first page; per-model identifiers are not offered.