O1 Mini (2024-09-12): benchmark results
September 12, 2024 O1 Mini snapshot, tracked when sources report the dated API model. Provider: OpenAI. Released 2024-09-12. Access: API.
Aggregate Rating 1611 ± 12, rank #403 of 1656 rated models, from 568 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| BrowserGym AssistantBench | 6.9 | Accuracy (%, GenericAgent unless named; tasks finished by ev | 100 |
| OpenCompass Code - Code QA - Chinese (CompassBench 2411) | 82.4 | Score (%) | 100 |
| OpenCompass Code - Code QA - Chinese (CompassBench 2501) | 81.9 | Score (%) | 100 |
| OpenCompass Code - Code QA - English (CompassBench 2411) | 82.8 | Score (%) | 100 |
| OpenCompass Code - Code QA - English (CompassBench 2501) | 80.6 | Score (%) | 100 |
| OpenCompass Code - Execution (CompassBench 2411) | 98 | Score (%) | 100 |
| OpenCompass Code - Generation (CompassBench 2411) | 56.4 | Score (%) | 100 |
| OpenCompass LLM - Average (CompassBench 2411) | 71.1 | Score (%) | 100 |
| OpenCompass LLM - Average - Chinese (CompassBench 2411) | 70.5 | Score (%) | 100 |
| OpenCompass LLM - Average - English (CompassBench 2411) | 73.2 | Score (%) | 100 |
| OpenCompass LLM - Code (CompassBench 2411) | 79.9 | Score (%) | 100 |
| OpenCompass LLM - Code - Chinese (CompassBench 2411) | 79 | Score (%) | 100 |
Interactive version: theaggregate.ai/model?slug=o1-mini-2024-09-12 · How It Works · Data refreshed daily, snapshot 2026-10-20.