Benchmark
SciCode
Coding benchmarks compare models on programming tasks. Scores are percentages. Higher is better.
- Models measured
- 154
- Scores recorded
- 265
- Sources
- 3
- IQ weight
- 1.71
Score distribution
- 47.7%
- Mean
- 50.1%
- Median
- 10.7%
- Std dev
- 35.4%
- 10th percentile
- 58.8%
- 90th percentile
- 1.3%
- Min
This benchmark has a weight of 1.71 for chat models, measured over 154 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 |
|---|---|---|
| 251 | Mistral Small2503· Epoch AI Benchmarking Hub data | 27.8% |
| 252 | Claude Haiku 3.5· Epoch AI Benchmarking Hub data | 27.4% |
| 253 | Mistral Small2506· Epoch AI Benchmarking Hub data | 26.4% |
| 254 | Llama 3.3 70B Instruct· Epoch AI Benchmarking Hub data | 26% |
| 255 | GPT-4.1 nano· Epoch AI Benchmarking Hub data | 25.9% |
| 256 | MiMo V2 Flash· Epoch AI Benchmarking Hub data | 25.9% |
| 257 | Solar Pro 3· Epoch AI Benchmarking Hub data | 25.5% |
| 258 | Granite 4.2 3B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 24.1% |
| 259 | Gemma 3 27B IT· Epoch AI Benchmarking Hub data | 23.3% |
| 260 | Qwen3 8B· Epoch AI Benchmarking Hub data | 22.6% |
| 261 | Llama 4 Scout· Epoch AI Benchmarking Hub data | 21.3% |
| 262 | Gemma 3 12B IT· Epoch AI Benchmarking Hub data | 16.4% |
| 263 | Llama 3.1 8B Instruct· Epoch AI Benchmarking Hub data | 13.2% |
| 264 | Phi 4 Mini· Epoch AI Benchmarking Hub data | 10.8% |
| 265 | Qwen3.6 35B A3Bnone· Epoch AI Benchmarking Hub data | 1.3% |
Showing 251–265 of 265 results.
Data credits: Data by LLM Stats · Epoch AI Benchmarking Hub, CC BY 4.0 (licence) · 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.