Benchmark
SEC-Bench Pro
Coding benchmarks compare models on programming tasks. Scores are percentages. Higher is better.
- Models measured
- 13
- Scores recorded
- 22
- Sources
- 2
- IQ weight
- 1.28
Score distribution
- 66%
- Mean
- 66.3%
- Median
- 15.6%
- Std dev
- 47.5%
- 10th percentile
- 86.8%
- 90th percentile
- 30.9%
- Min
This benchmark has a weight of 1.28 for chat models, measured over 13 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 | GPT-6.2 Sol· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 88.3% |
| 2 | Gemini 4 Carbon· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 88.2% |
| 3 | GPT-6.2 Astra· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 86.8% |
| 4 | GPT-6 Astra· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 85.4% |
| 5 | GPT-6 AstraOpenAI launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 85.4% |
| 6 | GPT-5.6 SolOpenAI launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 79.1% |
| 7 | GPT-5.6 SolDeepSeek V4.1 launch chart · exact public value · setting not further specified· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 74.3% |
| 8 | GPT-5.6 SolUltra effort· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 74.3% |
| 9 | GPT-5.6 Sol· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 71.2% |
| 10 | GPT-5.6 SolOpenAI GPT-5.6 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 71.2% |
| 11 | MiMo V2.6 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 66.3% |
| 12 | DeepSeek V4.1 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 62.8% |
| 13 | DeepSeek V4.1 FlashClaude Code harness· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 62.8% |
| 14 | DeepSeek V4.1 FlashDeepSeek V4.1 launch chart · exact public value · setting not further specified· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 62.8% |
| 15 | GPT-5.6 Terra· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 57.7% |
| 16 | GPT-5.6 TerraOpenAI GPT-5.6 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 57.7% |
| 17 | DeepSeek V4 Pro 0813DeepSeek V4.1 launch chart · exact public value · setting not further specified· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 56.4% |
| 18 | GPT-5.6 Luna· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 48.9% |
| 19 | GPT-5.6 LunaOpenAI GPT-5.6 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 48.9% |
| 20 | MiMo V2.6 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 47.5% |
| 21 | GPT 5.5OpenAI GPT-5.6 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 45.8% |
| 22 | DeepSeek V4 Flash 0731DeepSeek V4.1 launch chart · exact public value · setting not further specified· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 30.9% |
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