Qwen2.5-72B
ModelOpen sourceActiveby Alibaba (Qwen) · family “qwen2b”
Epoch Capabilities Index
129.0 #157 of 268
90% CI 123.0 – 130.8
Released
19 Sep 2024
Context window
Unknown
Max output
Unknown
Input $ / 1M tokens
Unknown
Output $ / 1M tokens
Unknown
Cached input $ / 1M
Unknown
Each benchmark shown separately with its own source
Benchmark results (18)
| Benchmark | Domain | Score | vs best recorded | Setting | Run | Source |
|---|---|---|---|---|---|---|
| LMCA | agents | 13.4% | 20% | — | — | External ↗ |
| DTBench | reasoning | 62.9% | 64% | — | — | External ↗ |
| METR Time Horizons | agents | 35.8% | 42% | — | — | External ↗ |
| GeoBench | multimodal | 62.0% | 70% | — | — | External ↗ |
| WeirdML | coding | 16.0% ±0.0 | 17% | — | — | External ↗ |
| OTIS Mock AIME 2024-2025 | math | 8.1% ±2.6 | 8% | — | 25 Feb 2025 | Eval log ↗ |
| The Agent Company | agents | 5.7% | 13% | — | — | External ↗ |
| Balrog | games | 16.2% | 24% | — | — | External ↗ |
| OSWorld | agents | 5.0% | 7% | agent: qwen2.5-vl-72b-instruct (100 steps) | — | External ↗ |
| GPQA diamond | science | 49.1% ±2.7 | 51% | — | 27 Jan 2025 | Eval log ↗ |
| BBH | reasoning | 79.8% | 89% | — | — | External ↗ |
| MATH level 5 | math | 63.2% ±1.1 | 64% | — | 27 Jan 2025 | Epoch ↗ |
| MMLU | knowledge | 85.3% | 97% | — | — | External ↗ |
| PIQA | other | 82.6% | 93% | — | — | External ↗ |
| Winogrande | other | 82.3% | 92% | — | — | External ↗ |
| HellaSwag | reasoning | 84.8% | 89% | — | — | External ↗ |
| ARC AI2 | other | 94.5% | 99% | — | — | External ↗ |
| TriviaQA | other | 71.9% | 82% | — | — | External ↗ |
Source: Epoch AI Benchmarking Hub (CC BY 4.0). “External” rows are leaderboard results Epoch collects from third parties. Best reported setting per benchmark is shown.
API list price over time
Price history
$ per 1M tokens
This model has no public API price listing
0 recorded prices on — (OpenRouter listing); re-read on every data refresh, most recently 11h ago. Steps show when the price changed.
Events
No events recorded yet.