Model comparison
o1 vs Qwen3 235B-A22B
Qwen3 235B-A22B is the stronger model overall, scoring 43.5 to 40.9 on the Noometry Index.
Last verified . 33 shared benchmarks.
Alibaba (Qwen)Summary
- They share 33 benchmarks with published results for both. o1 scores higher in 4 categories and Qwen3 235B-A22B in 5 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3 235B-A22B leads 50.4 to 36.1.
- The biggest single-benchmark swing is MATH Level 5: 94.7% for o1 and 68.9% for Qwen3 235B-A22B.
- Qwen3 235B-A22B is cheaper at $0.70 / $2.80 per million input/output tokens, against $15 / $60 for o1.
- o1 accepts more context: 200K tokens versus 131K.
- Qwen3 235B-A22B has downloadable open weights; the other is API-only.
Side by side
| o1 | Qwen3 235B-A22B | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 40.9 | 43.5 |
| Released | 2024-09-12 | 2025-04 |
| Weights | Proprietary | Open |
| Context window | 200K | 131K |
| Max output | 100K | 16K |
| Input $ / M tokens | $15 | $0.70 |
| Output $ / M tokens | $60 | $2.80 |
| Results tracked | 52 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o1 leads
o1: 46.1 (#70), Qwen3 235B-A22B: 44.3 (#75)
| Benchmark | o1 | Qwen3 235B-A22B |
|---|---|---|
| Aider Polyglot | 61.7% | 59.6% |
| WeirdML | 47.6% | 41% |
| LMArena Coding | 1367 | 1445 |
| SciCode | — | 42.4% |
| LiveBench Coding | 69.7% | — |
| CadEval | 56% | — |
| HumanEval+ | 89% | — |
| MBPP+ | 80.2% | — |
Agentic & Tool Use Qwen3 235B-A22B leads
o1: 24.6 (#117), Qwen3 235B-A22B: 33.9 (#51)
| Benchmark | o1 | Qwen3 235B-A22B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 52.1% |
| Cybench | 10% | — |
| METR Time Horizons | 51.1% | — |
| Vending-Bench 2 | — | -11.34 |
Reasoning o1 leads
o1: 27.9 (#111), Qwen3 235B-A22B: 15.7 (#311)
| Benchmark | o1 | Qwen3 235B-A22B |
|---|---|---|
| SimpleBench | 41.7% | 31% |
| ARC-AGI-1 | 30.7% | 11% |
| Chess Puzzles | 15% | 12% |
| LMArena Hard Prompts | 1371 | 1433 |
| DTBench | 74.7% | 80.3% |
| LMCA | 22.3% | 29.3% |
| Epoch Capabilities Index | 141.91 | 143.85 |
| ARC-AGI-2 | — | 1.3% |
| Kagi LLM Benchmark | — | 69.4% |
| CritPt | — | 0% |
| EnigmaEval | 5.7% | — |
| LiveBench Reasoning | 91.6% | — |
| Mystery Game Puzzles | — | 9% |
| LiveBench Data Analysis | 65.5% | — |
| ForecastBench | — | 59.7 |
| LiveBench | 75.7% | — |
Math Qwen3 235B-A22B leads
o1: 36.1 (#175), Qwen3 235B-A22B: 50.4 (#57)
| Benchmark | o1 | Qwen3 235B-A22B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.3% | 86.7% |
| LMArena Math | 1388 | 1432 |
| MATH Level 5 | 94.7% | 68.9% |
| FrontierMath (Feb 2025 set) | 9.3% | 8.5% |
| FrontierMath (Tiers 1-3) | 14.7% | — |
| Omni-MATH | — | 71.8% |
| LiveBench Math | 80.3% | — |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Qwen3 235B-A22B leads
o1: 41.5 (#110), Qwen3 235B-A22B: 49.6 (#73)
| Benchmark | o1 | Qwen3 235B-A22B |
|---|---|---|
| GPQA Diamond | 76.8% | 80.1% |
| SimpleQA Verified | 41.1% | 40.4% |
| Confabulations | 11.7% | 15.6% |
| LMArena Expert | 1361 | 1463 |
| Humanity's Last Exam | 8% | — |
| MMLU-Pro | — | 84.4% |
| Vectara Hallucination Rate | — | 9.3% |
| GPQA (HELM) | — | 72.7% |
Multimodal Not comparable
o1: 34.2 (#93), Qwen3 235B-A22B: —
| Benchmark | o1 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Vision | 1168 | — |
| GeoBench | 80% | — |
| VPCT | 37% | — |
| SpatialViz-Bench | 41.4% | — |
Multilingual Qwen3 235B-A22B leads
o1: 48.6 (#142), Qwen3 235B-A22B: 52.3 (#89)
| Benchmark | o1 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Non-English | 1358 | 1409 |
| LMArena Chinese | 1394 | 1481 |
| LMArena French | 1344 | 1445 |
| LMArena German | 1337 | 1433 |
| LMArena Japanese | 1346 | 1399 |
| LMArena Korean | 1396 | 1391 |
| LMArena Russian | 1356 | 1411 |
| LMArena Spanish | 1345 | 1430 |
Instruction Following o1 leads
o1: 74.8 (#86), Qwen3 235B-A22B: 72.6 (#136)
| Benchmark | o1 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Instruction Following | 1367 | 1408 |
| LiveBench Instruction Following | 81.5% | — |
| IFEval | — | 83.5% |
Long Context o1 leads
o1: 50.3 (#9), Qwen3 235B-A22B: 46.1 (#26)
| Benchmark | o1 | Qwen3 235B-A22B |
|---|---|---|
| Fiction.LiveBench | 83.3% | 75% |
| LMArena Longer Query | 1378 | 1426 |
Writing & Preference Qwen3 235B-A22B leads
o1: 55.6 (#144), Qwen3 235B-A22B: 59.6 (#108)
| Benchmark | o1 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Text | 1366 | 1419 |
| LMArena Creative Writing | 1348 | 1384 |
| Short-Story Creative Writing | 70.2% | 83% |
| LMArena Multi-Turn | 1369 | 1432 |
| EQ-Bench Creative Writing | — | 1366 |
| WildBench | — | 86.6% |
| LiveBench Language | 65.4% | — |
Frequently asked questions
Is o1 better than Qwen3 235B-A22B?
Qwen3 235B-A22B is the stronger model overall, scoring 43.5 to 40.9 on the Noometry Index.
Which is cheaper, o1 or Qwen3 235B-A22B?
Qwen3 235B-A22B is cheaper. It lists at $0.70 per million input tokens and $2.80 per million output tokens; o1 lists at $15 and $60.
Is o1 or Qwen3 235B-A22B better for coding?
o1 scores higher on coding benchmarks: 46.1 versus 44.3 in the Noometry coding category.
Which has the bigger context window?
o1 does, with 200K tokens against 131K.
How many benchmarks do o1 and Qwen3 235B-A22B share?
33 benchmarks have published results for both models. o1 has 52 scored results on Noometry and Qwen3 235B-A22B has 49.