developer_tools
Best LLM for Debugging
Find the top-ranked models for localizing bugs and proposing fixes with explanations.
Full Analysis Available
Benchmark methodology, patterns in the data, and deployment notes
Provisional leader
anthropic/claude-sonnet-4
Best current option from the available benchmark evidence, but not yet a strong winner claim.
external/anthropic/claude-sonnet-4
29.8%
Score
41.7%
Confidence
27
Evidence
$6.00
per 1M tokens
Ranked Models
30
Evidence Quality
81%
Evidence Points
27
Top Signal
SWE-bench Verified Leaderboard: swe_verified_resolved_pct
All Ranked Models
| Rank | Model | Score |
|---|---|---|
| 🥇 | claude-sonnet-4 Strong on SWE-bench Verified Leaderboard swe_verified_resolved_pct and Galileo Agent Leaderboard v2 Avg AC | 29.8% |
| 🥈 | gpt-4o-2024-05-13 Strong on RepoQA Official Results overall_average_pass_at_1_pct and SWE-bench Verified Leaderboard swe_verified_resolved_pct | 27.2% |
| 🥉 | gpt-5-2025-08-07 Strong on SWE-bench Verified Leaderboard swe_verified_resolved_pct and Aider Polyglot Leaderboard percent_correct_pct | 26.0% |
| #6 | Kimi K2 Thinking Strong on SWE-bench Verified Leaderboard swe_verified_resolved_pct and Sonar Java Quality Leaderboard functional_skill_pct | 23.3% |
| #8 | gemini-3-pro-preview Strong on SWE-bench Verified Leaderboard swe_verified_resolved_pct and Vals SWE-bench overall_accuracy_pct | 19.9% |
| #9 | gemini-2.5-pro Strong on SWE-bench Verified Leaderboard swe_verified_resolved_pct and Galileo Agent Leaderboard v2 Avg AC | 19.9% |
| #10 | gpt-5.2-2025-12-11 Strong on SWE-bench Verified Leaderboard swe_verified_resolved_pct and Vals SWE-bench overall_accuracy_pct | 19.8% |
| #12 | o3-20250416 Strong on SWE-bench Verified Leaderboard swe_verified_resolved_pct and Aider Polyglot Leaderboard percent_correct_pct | 18.9% |
| #13 | gpt-5-mini-2025-08-07 Strong on SWE-bench Verified Leaderboard swe_verified_resolved_pct and Vals LiveCodeBench overall_accuracy_pct | 17.9% |
| #14 | glm-4.7 Strong on Sonar Java Quality Leaderboard functional_skill_pct and Vals LiveCodeBench overall_accuracy_pct | 17.6% |
| #15 | deepseek-r1 Strong on Sonar Java Quality Leaderboard functional_skill_pct and Aider Polyglot Leaderboard percent_correct_pct | 17.2% |
| #17 | claude-opus-4-5-20251101 Strong on SWE-bench Verified Leaderboard swe_verified_resolved_pct and Vals Terminal-Bench 2 overall_accuracy_pct | 17.1% |
| #18 | gpt-4.1-20250414 Strong on Galileo Agent Leaderboard v2 Avg AC and SWE-bench Verified Leaderboard swe_verified_resolved_pct | 16.9% |
| #20 | minimax-m2.1 Strong on Sonar Java Quality Leaderboard functional_skill_pct and Vals SWE-bench overall_accuracy_pct | 16.9% |
| #21 | gpt-4o-20241120 Strong on Aider Code Editing Leaderboard percent_correct_pct and BigCodeBench Official bigcodebench_complete_pct | 16.6% |
| #22 | gemini-3.1-pro-preview Strong on Vals SWE-bench overall_accuracy_pct and Vals Terminal-Bench 2 overall_accuracy_pct | 16.1% |
| #23 | claude-opus-4 Strong on SWE-bench Verified Leaderboard swe_verified_resolved_pct and Aider Polyglot Leaderboard percent_correct_pct | 15.7% |
| #25 | o4-mini Strong on SWE-bench Verified Leaderboard swe_verified_resolved_pct and Aider Polyglot Leaderboard percent_correct_pct | 15.1% |
| #27 | qwen-2.5-72b-instruct Strong on Galileo Agent Leaderboard v2 Avg AC and Aider Code Editing Leaderboard percent_correct_pct | 14.6% |
| #28 | gpt-4.1 Strong on SWE-bench Verified Leaderboard swe_verified_resolved_pct and Aider Polyglot Leaderboard percent_correct_pct | 14.5% |
| #30 | Grok-4-0709 Strong on Vals LiveCodeBench overall_accuracy_pct and Galileo Agent Leaderboard v2 Avg AC | 14.2% |
| #31 | gpt-4o Strong on Sonar Java Quality Leaderboard functional_skill_pct and SWE-bench Verified Leaderboard swe_verified_resolved_pct | 13.9% |
| #34 | gemini-3-flash-preview Strong on Vals SWE-bench overall_accuracy_pct and Vals LiveCodeBench overall_accuracy_pct | 13.5% |
| #37 | gpt-5.4-2026-03-05 Strong on Vals SWE-bench overall_accuracy_pct and Vals LiveCodeBench overall_accuracy_pct | 13.5% |
| #38 | Phi-3-medium-128k-instruct Strong on RepoQA Official Results overall_average_pass_at_1_pct and RepoQA Official Results all_average_pass_at_1_pct | 13.3% |
| #39 | claude-sonnet-4.6 Strong on Vals SWE-bench overall_accuracy_pct and Vals LiveCodeBench overall_accuracy_pct | 13.1% |
| #40 | claude-opus-4-6-thinking Strong on Vals SWE-bench overall_accuracy_pct and Vals LiveCodeBench overall_accuracy_pct | 12.9% |
| #43 | GLM-5 Strong on Sonar Java Quality Leaderboard functional_skill_pct and Sonar Java Quality Leaderboard issue_density_error_per_kloc | 12.7% |
| #45 | gpt-5.1-2025-11-13 Strong on Vals LiveCodeBench overall_accuracy_pct and Vals SWE-bench overall_accuracy_pct | 12.4% |
| #46 | claude-opus-4-5-20251101-thinking Strong on Vals SWE-bench overall_accuracy_pct and Vals LiveCodeBench overall_accuracy_pct | 12.4% |
Head-to-Head: #1 vs #2
#1
Top Pickanthropic/claude-sonnet-4
Strong on SWE-bench Verified Leaderboard swe_verified_resolved_pct and Galileo Agent Leaderboard v2 Avg AC
Conf 41.7%
#2
gpt-4o-2024-05-13
Strong on RepoQA Official Results overall_average_pass_at_1_pct and SWE-bench Verified Leaderboard swe_verified_resolved_pct
Conf 35.5%
Related Lookups
Best LLM for Code Generation
Benchmark-backed ranking of models for generating correct, secure code from requirements.
Best LLM for Unit Test Generation
Ranked models for generating meaningful unit tests and edge cases from code.
Best LLM for Code Review
Compare models for automated PR review covering correctness, security, and maintainability.
Best LLM for Autonomous Coding
Benchmark-backed ranking of models for end-to-end autonomous software engineering and issue resolution.
Best LLM for Function Calling
Compare models for reliable tool use, function selection, and multi-step API orchestration.
Best LLM for Refactoring
Ranked models for safely refactoring code while preserving behavior and improving clarity.