Model comparison
o3 vs Qwen3.5 122B-A10B
o3 is the stronger model overall, scoring 47.4 to 41.6 on the Noometry Index. Qwen3.5 122B-A10B costs 3.2× less per token, which makes it the better buy when o3's lead doesn't matter for your workload.
Last verified . 22 shared benchmarks.
Alibaba (Qwen)Summary
- They share 22 benchmarks with published results for both. o3 scores higher in 8 categories and Qwen3.5 122B-A10B in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where o3 leads 54.5 to 38.2.
- The biggest single-benchmark swing is Mystery Game Puzzles: 29% for o3 and 17% for Qwen3.5 122B-A10B.
- Qwen3.5 122B-A10B is cheaper at $0.40 / $3.20 per million input/output tokens, against $2 / $8 for o3.
- Qwen3.5 122B-A10B accepts more context: 262K tokens versus 200K.
- Qwen3.5 122B-A10B has downloadable open weights; the other is API-only.
Side by side
| o3 | Qwen3.5 122B-A10B | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 47.4 | 41.6 |
| Released | 2025-04-16 | 2026-02-23 |
| Weights | Proprietary | Open |
| Context window | 200K | 262K |
| Max output | 100K | 66K |
| Input $ / M tokens | $2 | $0.40 |
| Output $ / M tokens | $8 | $3.20 |
| Results tracked | 63 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3 leads
o3: 47.3 (#57), Qwen3.5 122B-A10B: 38.8 (#155)
| Benchmark | o3 | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Coding | 1408 | 1436 |
| SWE-bench Verified | 62.3% | — |
| SWE-bench Verified (bash only) | 58.4% | — |
| Aider Polyglot | 81.3% | — |
| LMArena WebDev | — | 1360 |
| SciCode | — | 35.6% |
| GSO | 8.8% | — |
| WeirdML | 52.4% | — |
| CadEval | 74% | — |
| ALE-Bench | 933.55 | — |
Agentic & Tool Use Not comparable
o3: 34.3 (#42), Qwen3.5 122B-A10B: —
| Benchmark | o3 | Qwen3.5 122B-A10B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 63% | — |
| GDPval | 30.8% | — |
| DeepResearch Bench | 45.2% | — |
| OSWorld | 23% | — |
| LMArena Search | 1144 | — |
| METR Time Horizons | 65.4% | — |
Reasoning o3 leads
o3: 32.1 (#66), Qwen3.5 122B-A10B: 26.4 (#121)
| Benchmark | o3 | Qwen3.5 122B-A10B |
|---|---|---|
| CritPt | 1.4% | 0.9% |
| LMArena Hard Prompts | 1402 | 1421 |
| Mystery Game Puzzles | 29% | 17% |
| DTBench | 84.8% | 84.3% |
| LMCA | 39.7% | 32.2% |
| ARC-AGI-2 | 6.5% | — |
| SimpleBench | 53.1% | — |
| Kagi LLM Benchmark | 67.6% | — |
| NYT Connections (extended) | — | 51.7% |
| ARC-AGI-1 | 60.8% | — |
| Chess Puzzles | 38% | — |
| EnigmaEval | 13.1% | — |
| Thematic Generalization | — | 51.2% |
| Epoch Capabilities Index | 146.88 | — |
| ForecastBench | 62.5 | — |
Math o3 leads
o3: 48.9 (#49), Qwen3.5 122B-A10B: 36.4 (#114)
| Benchmark | o3 | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Math | 1426 | 1432 |
| FrontierMath (Tiers 1-3) | 33.3% | — |
| OTIS Mock AIME 2024-2025 | 84.4% | — |
| Omni-MATH | 71.4% | — |
| MATH Level 5 | 97.8% | — |
| FrontierMath (Feb 2025 set) | 18.7% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge o3 leads
o3: 54.5 (#47), Qwen3.5 122B-A10B: 38.2 (#138)
| Benchmark | o3 | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Expert | 1403 | 1431 |
| GPQA Diamond | 81.8% | — |
| Humanity's Last Exam | 20.3% | — |
| SimpleQA Verified | 49.4% | — |
| MMLU-Pro | 85.9% | — |
| Confabulations | 14.4% | — |
| Vectara Hallucination Rate | — | 11.2% |
| GPQA (HELM) | 75.3% | — |
Multimodal o3 leads
o3: 40.7 (#35), Qwen3.5 122B-A10B: 39.0 (#52)
| Benchmark | o3 | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Vision | 1214 | 1245 |
| GeoBench | 74% | — |
| VPCT | 52% | — |
Multilingual Too close to call
o3: 51.9 (#99), Qwen3.5 122B-A10B: 51.8 (#101)
| Benchmark | o3 | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Non-English | 1402 | 1400 |
| LMArena Chinese | 1438 | 1462 |
| LMArena French | 1430 | 1442 |
| LMArena German | 1420 | 1425 |
| LMArena Japanese | 1404 | 1364 |
| LMArena Korean | 1371 | 1353 |
| LMArena Russian | 1407 | 1400 |
| LMArena Spanish | 1395 | 1423 |
Instruction Following Qwen3.5 122B-A10B leads
o3: 73.0 (#122), Qwen3.5 122B-A10B: 74.0 (#109)
| Benchmark | o3 | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Instruction Following | 1368 | 1399 |
| IFEval | 86.9% | — |
Long Context o3 leads
o3: 53.2 (#6), Qwen3.5 122B-A10B: 43.0 (#105)
| Benchmark | o3 | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Longer Query | 1372 | 1410 |
| Fiction.LiveBench | 88.9% | — |
| CL-bench | 17.8% | — |
Writing & Preference o3 leads
o3: 63.7 (#62), Qwen3.5 122B-A10B: 60.2 (#98)
| Benchmark | o3 | Qwen3.5 122B-A10B |
|---|---|---|
| LMArena Text | 1410 | 1417 |
| LMArena Creative Writing | 1359 | 1369 |
| LMArena Multi-Turn | 1405 | 1416 |
| Short-Story Creative Writing | 83.9% | — |
| EQ-Bench Creative Writing | 1676 | — |
| WildBench | 86.1% | — |
Frequently asked questions
Is o3 better than Qwen3.5 122B-A10B?
o3 is the stronger model overall, scoring 47.4 to 41.6 on the Noometry Index. Qwen3.5 122B-A10B costs 3.2× less per token, which makes it the better buy when o3's lead doesn't matter for your workload.
Which is cheaper, o3 or Qwen3.5 122B-A10B?
Qwen3.5 122B-A10B is cheaper. It lists at $0.40 per million input tokens and $3.20 per million output tokens; o3 lists at $2 and $8.
Is o3 or Qwen3.5 122B-A10B better for coding?
o3 scores higher on coding benchmarks: 47.3 versus 38.8 in the Noometry coding category.
Which has the bigger context window?
Qwen3.5 122B-A10B does, with 262K tokens against 200K.
How many benchmarks do o3 and Qwen3.5 122B-A10B share?
22 benchmarks have published results for both models. o3 has 63 scored results on Noometry and Qwen3.5 122B-A10B has 27.