ENAMEL: leaderboard
Efficiency-aware code-generation benchmark built from HumanEval problems with expert efficient reference solutions and strong test generators, reporting eff@1 alongside pass@1.
Metric: Greedy eff@1 (%) (imported). Source: huggingface.co. Source: benchmarklist.com. Saturation forecast: Estimated already saturated. 30 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4 Preview (1106) | 47 | #418 |
| 2 | GPT-4 (0613) | 45.4 | #491 |
| 3 | Llama 3 70B Instruct | 42.1 | #634 |
| 4 | Mixtral 8x22B Instruct | 40.8 | #728 |
| 5 | Claude 3 Opus (20240229) | 40.1 | #412 |
| 6 | Claude 3 Haiku (20240307) | 38.6 | #771 |
| 7 | ChatGPT | 36.4 | #900 |
| 8 | Claude 3 Sonnet (20240229) | 34.5 | #677 |
| 9 | Llama 3 8B Instruct | 34.4 | #1129 |
| 10 | Mixtral 8x7B Instruct | 26.6 | #1170 |
| 11 | starcoder | 19.5 | #1577 |
| 12 | Mistral 7B | 15.2 | #1397 |
| 13 | vicuna-13B | 12.3 | #1469 |
| 14 | Vicuna-7B | 6.1 | #1542 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=enamel · How It Works · Data refreshed daily, snapshot 2026-10-20.