o1vsQwen2.5
o1 | Qwen2.5 | |
|---|---|---|
| Specifications | ||
Parameters | — | 72B |
Context window | 200k | 128k |
| Benchmarks | ||
Science GPQA DiamondGraduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 75.7% | — |
| Overview | ||
| Company | OpenAI | Qwen |
| Release date | Dec 5 2024 | Sep 19 2024 |
| Access | Proprietary | Open Weight |
Which is better: o1 or Qwen2.5?
o1 and Qwen2.5 don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. Qwen2.5 shipped 77 days before o1, so benchmark comparisons should account for the intervening progress.
Context windows are 200k (o1) vs 128k (Qwen2.5). o1 is proprietary, while Qwen2.5 is open weight.
Direct benchmark comparisons are unavailable — o1 and Qwen2.5 don't publish scores on any of the same benchmarks.
Frequently asked questions
o1 was released by OpenAI on Dec 5 2024.
Qwen2.5 was released by Qwen on Sep 19 2024.
o1 has a 200k context window; Qwen2.5 has 128k.
o1 is a proprietary model released by OpenAI. Qwen2.5 is an open weight model released by Qwen.