Home / Best AI for… / Best AI for Coding
Task ranking
Best AI for Coding
Ranked by SWE-bench Verified — real GitHub issues resolved end-to-end — with League Score as the tiebreak.
Our top pick: Claude Opus 5 — #1 on SWE-bench Verified (96.0%) · League Score 92 · $5.00/$25.00 per 1M. Full record, pricing and benchmarks on its model page; or run the model picker for a recommendation matched to your budget and context needs.
| # | Model | Maker | Why it ranks here |
|---|---|---|---|
| 1 | Claude Opus 5 | Anthropic | #1 on SWE-bench Verified (96.0%) · League Score 92 · $5.00/$25.00 per 1M |
| 2 | Claude Sonnet 5 | Anthropic | #2 on SWE-bench Verified (85.2%) · League Score 74 · $2.00/$10.00 per 1M |
| 3 | Kimi K2.6 | Moonshot AI | #3 on SWE-bench Verified (80.2%) · League Score 70 · $0.95/$4.00 per 1M |
| 4 | Mistral Medium 3.5 | Mistral AI | #4 on SWE-bench Verified (77.6%) · League Score 59 · $1.50/$7.50 per 1M |
| 5 | Qwen3.5-397B-A17B | Alibaba Qwen | #5 on SWE-bench Verified (76.2%) · League Score 61 |
| 6 | Gemini 3.1 Pro Preview | Google DeepMind | #6 on SWE-bench Verified (75.6%) · League Score 52 · $2.00/$12.00 per 1M |
| 7 | GPT-5 | OpenAI | #7 on SWE-bench Verified (73.6%) · League Score 60 · $1.25/$10.00 per 1M |
| 8 | DeepSeek V4 Pro (0813) | DeepSeek | #8 on SWE-bench Verified (73.6%) · League Score 63 · $1.32/$3.96 per 1M |
| 9 | Claude Haiku 4.5 | Anthropic | #9 on SWE-bench Verified (73.3%) · League Score 16 · $1.00/$5.00 per 1M |
| 10 | Nemotron 3 Ultra 550B-A55B | NVIDIA | #10 on SWE-bench Verified (71.9%) · League Score 57 |
Rankings regenerate from the model database on every data update. Metrics and sources: methodology.