Benchmark
SWE-bench Verified
Coding benchmarks compare models on programming tasks. Scores are percentages. Higher is better.
- Models measured
- 130
- Scores recorded
- 205
- Sources
- 5
- IQ weight
- 2.70
Score distribution
- 70%
- Mean
- 73.5%
- Median
- 13.7%
- Std dev
- 51.6%
- 10th percentile
- 80.8%
- 90th percentile
- 8.7%
- Min
This benchmark has a weight of 2.70 for chat models, measured over 129 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 5· Steel.dev Agent Leaderboards data | 96% |
| 2 | Claude Fable 5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 95% |
| 3 | Claude Fable 5Fable/Mythos 5 System Card capability summary · adaptive thinking · max effort · 5 trials unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 95% |
| 4 | Ember-1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 92.2% |
| 5 | Claude Opus 4.8· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 88.6% |
| 6 | Claude Opus 4.8Fable/Mythos 5 System Card capability summary · adaptive thinking · max effort · 5 trials unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 88.6% |
| 7 | Claude Opus 4.8Opus 4.8 System Card capability summary · adaptive thinking · max effort · 5 trials unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 88.6% |
| 8 | Claude Opus 4.7· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 87.6% |
| 9 | Claude Opus 4.7adaptive thinking · max effort · avg@5· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 87.6% |
| 10 | Claude Opus 4.7Opus 4.8 System Card capability summary · adaptive thinking · max effort · 5 trials unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 87.6% |
| 11 | Claude Sonnet 5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 85.2% |
| 12 | Claude Sonnet 5adaptive thinking · max effort · 5 trials· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 85.2% |
| 13 | GPT-5.6 Sol· Steel.dev Agent Leaderboards data | 82.2% |
| 14 | Inkling 2 Small· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 82.2% |
| 15 | Claude Sonnet 4.5parallel test-time compute· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 82% |
| 16 | Beam 1.0 501B A23B· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 81.6% |
| 17 | Claude Opus 4.6Anthropic Opus 4.6 launch page · adaptive thinking · max effort unless noted · prompt modification · 25 trials· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 81.4% |
| 18 | Claude Opus 4.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 80.9% |
| 19 | Claude Opus 4.5no extended thinking · 200K context · avg@5· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 80.9% |
| 20 | Claude Opus 4.5Opus 4.6 System Card capability summary · adaptive thinking · max effort · default sampling · 5 trials unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 80.9% |
| 21 | Claude Opus 4.6· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 80.8% |
| 22 | Claude Opus 4.6Opus 4.6 System Card capability summary · adaptive thinking · max effort · default sampling · 5 trials unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 80.8% |
| 23 | Claude Opus 4.564K thinking budget · 200K context · avg@5· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 80.6% |
| 24 | DeepSeek V4 Pro· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim | 80.6% |
| 25 | Gemini 3.1 Pro Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 80.6% |
| 26 | Gemini 3.1 Pro PreviewFable/Mythos 5 System Card capability summary · adaptive thinking · max effort · 5 trials unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 80.6% |
| 27 | Gemini 3.1 Pro PreviewOpus 4.8 System Card capability summary · adaptive thinking · max effort · 5 trials unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 80.6% |
| 28 | Gemini 3.1 Pro Previewsingle attempt· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 80.6% |
| 29 | MiniMax M3· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 80.5% |
| 30 | MiniMax M3Claude Code · four-run mean· Reported by MiniMax Research· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 80.5% |
| 31 | Qwen3.7 Max· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 80.4% |
| 32 | Qwen3.7 MaxQwen3.7-Max release · internal bash + file-edit agent scaffold · temp 1.0 · top_p 0.95 · 200K· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 80.4% |
| 33 | Claude Sonnet 4parallel test-time compute · regression-test rejection · internal scoring model selects one patch· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 80.2% |
| 34 | Claude Sonnet 4.6prompt-modified ablation · >100 tool calls · own tests first· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 80.2% |
| 35 | Inkling Small· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 80.2% |
| 36 | Kimi K2.6· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 3 sources
| 80.2% |
| 37 | Kimi K2.6Kimi K2.6 official benchmark table · thinking · temp 1.0 · top_p 1.0 · 262,144 context · in-house SWE-agent-derived harness · avg@10· Reported by Moonshot AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 80.2% |
| 38 | MiniMax M2.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 4 sources
| 80.2% |
| 39 | MiniMax M2.5main capability figure· Reported by MiniMax· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 80.2% |
| 40 | GLM 5.2· Steel.dev Agent Leaderboards data | 80% |
| 41 | GPT 5.2· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 80% |
| 42 | GPT 5.2Opus 4.6 System Card capability summary · adaptive thinking · max effort · default sampling · 5 trials unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 80% |
| 43 | GPT 5.2xhigh· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 80% |
| 44 | MiniMax M2.5official prose · Droid · default prompt· Reported by MiniMax· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 79.7% |
| 45 | Claude Sonnet 4.6· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 79.6% |
| 46 | Claude Sonnet 4.6adaptive thinking · max effort · 10 trials· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 79.6% |
| 47 | Claude Opus 4parallel test-time compute · regression-test rejection · internal scoring model selects one patch· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 79.4% |
| 48 | Inkling 2 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 79.2% |
| 49 | DeepSeek V4 Flash· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 79% |
| 50 | DeepSeek V4 FlashDeepSeek V4 official model card · Think Max· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 79% |
Showing 1–50 of 205 results.
Data credits: Data by LLM Stats · Hugging Face official benchmark leaderboards; licence varies by benchmark · BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · Steel.dev Agent Leaderboards, MIT (licence) · OpenEvals/leaderboard-data, MIT (licence)