Benchmark
FrontierSWE
Coding benchmarks compare models on programming tasks. Scores are percentages. Higher is better.
- Models measured
- 16
- Scores recorded
- 22
- Sources
- 2
- IQ weight
- 1.33
Score distribution
- 62.3%
- Mean
- 73.5%
- Median
- 21.1%
- Std dev
- 27%
- 10th percentile
- 81.2%
- 90th percentile
- 22%
- Min
This benchmark has a weight of 1.33 for chat models, measured over 16 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 Fable 5· LLM Stats data | 90% |
| 2 | Claude Fable 5Qwen表:Claude Code;对照为 2026-08-03 官方榜 MEAN@5· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 88.8% |
| 3 | Kimi K3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 81.2% |
| 4 | Kimi K3Kimi Code · Dominance · snapshot 2026-07-16· Reported by Moonshot AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 81.2% |
| 5 | GLM 5.3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 78.1% |
| 6 | GLM 5.3Dominance · Proximal · 1M · max effort · 128K output · snapshot 2026-08-14· Reported by Z.ai· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 78.1% |
| 7 | Claude Opus 4.8· LLM Stats data | 75% |
| 8 | GLM 5.2Dominance · 1M context · max effort · 128K output · as of 2026-06-16· Reported by Z.ai· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 74.4% |
| 9 | GLM 5.2· LLM Stats data | 74% |
| 10 | Qwen3.8 Max· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 73.5% |
| 11 | Qwen3.8 MaxQwen表:Claude Code;对照为 2026-08-03 官方榜 MEAN@5· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 73.5% |
| 12 | GPT 5.5· LLM Stats data | 73% |
| 13 | Claude Opus 4.8Qwen表:Claude Code;对照为 2026-08-03 官方榜 MEAN@5· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 70% |
| 14 | Claude Opus 4.7· LLM Stats data | 63% |
| 15 | Claude Opus 4.6· LLM Stats data | 56% |
| 16 | GPT 5.4· LLM Stats data | 54% |
| 17 | Qwen3.7 MaxQwen表:Claude Code;对照为 2026-08-03 官方榜 MEAN@5· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 40.7% |
| 18 | Gemini 3.1 Pro Preview· LLM Stats data | 40% |
| 19 | GLM 5.1· LLM Stats data | 31% |
| 20 | Kimi K2.6· LLM Stats data | 27% |
| 21 | Kimi K2.5· LLM Stats data | 26% |
| 22 | Qwen3.6 Plus· LLM Stats data | 22% |
Data credits: 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. · Data by LLM Stats