Model comparison
o3 vs Qwen3 Max
o3 is the stronger model overall, scoring 47.5 to 43.7 on the Noometry Index.
Last verified . 31 shared benchmarks.
Alibaba (Qwen)Summary
- They share 31 benchmarks with published results for both. o3 scores higher in 6 categories and Qwen3 Max in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in long context, where o3 leads 53.3 to 41.6.
- The biggest single-benchmark swing is Chess Puzzles: 38% for o3 and 4% for Qwen3 Max.
- Qwen3 Max is cheaper at $1.20 / $6 per million input/output tokens, against $2 / $8 for o3.
- Qwen3 Max accepts more context: 262K tokens versus 200K.
Side by side
| o3 | Qwen3 Max | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 47.5 | 43.7 |
| Released | 2025-04-16 | 2025-09-23 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | 262K |
| Max output | 100K | 66K |
| Input $ / M tokens | $2 | $1.20 |
| Output $ / M tokens | $8 | $6 |
| Results tracked | 63 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3 leads
o3: 46.8 (#64), Qwen3 Max: 43.0 (#93)
| Benchmark | o3 | Qwen3 Max |
|---|---|---|
| LMArena Coding | 1408 | 1456 |
| ALE-Bench | 933.55 | 370.45 |
| SWE-bench Verified | 62.3% | — |
| SWE-bench Verified (bash only) | 58.4% | — |
| Aider Polyglot | 81.3% | — |
| GSO | 8.8% | — |
| WeirdML | 52.4% | — |
| CadEval | 74% | — |
Agentic & Tool Use Not comparable
o3: 34.5 (#44), Qwen3 Max: —
| Benchmark | o3 | Qwen3 Max |
|---|---|---|
| Berkeley Function Calling Leaderboard | 63% | — |
| GDPval | 30.8% | — |
| DeepResearch Bench | 45.2% | — |
| OSWorld | 23% | — |
| LMArena Search | 1144 | — |
| METR Time Horizons | 65.4% | — |
| Vending-Bench 2 | — | 71.56 |
Reasoning o3 leads
o3: 32.0 (#78), Qwen3 Max: 22.6 (#190)
| Benchmark | o3 | Qwen3 Max |
|---|---|---|
| Kagi LLM Benchmark | 67.6% | 72.5% |
| Chess Puzzles | 38% | 4% |
| LMArena Hard Prompts | 1402 | 1448 |
| Mystery Game Puzzles | 29% | 5% |
| DTBench | 84.8% | 82.1% |
| LMCA | 39.7% | 28.3% |
| Epoch Capabilities Index | 146.86 | 142.38 |
| ARC-AGI-2 | 6.5% | — |
| SimpleBench | 53.1% | — |
| NYT Connections (extended) | — | 30.1% |
| ARC-AGI-1 | 60.8% | — |
| CritPt | 1.4% | — |
| EnigmaEval | 13.1% | — |
| ForecastBench | 62.5 | — |
Math o3 leads
o3: 50.2 (#58), Qwen3 Max: 38.7 (#131)
| Benchmark | o3 | Qwen3 Max |
|---|---|---|
| FrontierMath (Tiers 1-3) | 33.3% | 18.9% |
| OTIS Mock AIME 2024-2025 | 84.4% | 73.3% |
| LMArena Math | 1426 | 1446 |
| MATH Level 5 | 97.8% | 97.1% |
| Omni-MATH | 71.4% | — |
| FrontierMath (Feb 2025 set) | 18.7% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge o3 leads
o3: 54.6 (#52), Qwen3 Max: 48.1 (#78)
| Benchmark | o3 | Qwen3 Max |
|---|---|---|
| GPQA Diamond | 81.8% | 72.6% |
| SimpleQA Verified | 49.4% | 48.7% |
| LMArena Expert | 1402 | 1455 |
| Humanity's Last Exam | 20.3% | — |
| MMLU-Pro | 85.9% | — |
| Confabulations | 14.4% | — |
| GPQA (HELM) | 75.3% | — |
Multimodal Not comparable
o3: 41.4 (#36), Qwen3 Max: —
| Benchmark | o3 | Qwen3 Max |
|---|---|---|
| LMArena Vision | 1214 | — |
| GeoBench | 74% | — |
| VPCT | 52% | — |
Multilingual Qwen3 Max leads
o3: 51.7 (#105), Qwen3 Max: 53.7 (#62)
| Benchmark | o3 | Qwen3 Max |
|---|---|---|
| LMArena Non-English | 1401 | 1429 |
| LMArena Chinese | 1437 | 1478 |
| LMArena French | 1430 | 1449 |
| LMArena German | 1420 | 1463 |
| LMArena Japanese | 1403 | 1397 |
| LMArena Korean | 1370 | 1399 |
| LMArena Russian | 1406 | 1428 |
| LMArena Spanish | 1395 | 1462 |
Instruction Following Qwen3 Max leads
o3: 72.8 (#127), Qwen3 Max: 74.8 (#87)
| Benchmark | o3 | Qwen3 Max |
|---|---|---|
| LMArena Instruction Following | 1368 | 1419 |
| IFEval | 86.9% | — |
Long Context o3 leads
o3: 53.3 (#6), Qwen3 Max: 41.6 (#134)
| Benchmark | o3 | Qwen3 Max |
|---|---|---|
| Fiction.LiveBench | 88.9% | 66.7% |
| CL-bench | 17.8% | 14.5% |
| LMArena Longer Query | 1372 | 1438 |
Writing & Preference o3 leads
o3: 63.5 (#64), Qwen3 Max: 62.4 (#76)
| Benchmark | o3 | Qwen3 Max |
|---|---|---|
| LMArena Text | 1410 | 1439 |
| LMArena Creative Writing | 1359 | 1402 |
| LMArena Multi-Turn | 1405 | 1446 |
| Short-Story Creative Writing | 83.9% | — |
| EQ-Bench Creative Writing | 1676 | — |
| WildBench | 86.1% | — |
Frequently asked questions
Is o3 better than Qwen3 Max?
o3 is the stronger model overall, scoring 47.5 to 43.7 on the Noometry Index.
Which is cheaper, o3 or Qwen3 Max?
Qwen3 Max is cheaper. It lists at $1.20 per million input tokens and $6 per million output tokens; o3 lists at $2 and $8.
Is o3 or Qwen3 Max better for coding?
o3 scores higher on coding benchmarks: 46.8 versus 43.0 in the Noometry coding category.
Which has the bigger context window?
Qwen3 Max does, with 262K tokens against 200K.
How many benchmarks do o3 and Qwen3 Max share?
31 benchmarks have published results for both models. o3 has 63 scored results on Noometry and Qwen3 Max has 33.