Up to four models
o3 vs Qwen3-8B
On the index (GPT-4 = 100), o3 scores 147 and Qwen3-8B 126. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
o3 leads Qwen3-8B by 22 points, ahead on 1 of 1 shared core tests.
Test by testthird-party results
o3
Qwen3-8B
Every result as a table
| Test | o3 | Qwen3-8B |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 81.8% | 56.8% |
| Fiction.LiveBench | 88.9% | 62.1% |
| Maths | ||
| AIME-style maths (OTIS mock) | 84.4% | 56.1% |
| Agents | ||
| DTBench | 84.8% | 59.7% |
| LMCA | 39.7% | 8.8% |
Task length · METR39 sec
Tests at least two of these models have taken. Bold rows are the core tests.