Model comparison
o1 vs Qwen3.5 397B-A17B
Qwen3.5 397B-A17B is the stronger model overall, scoring 45.7 to 40.6 on the Noometry Index.
Last verified . 25 shared benchmarks.
Alibaba (Qwen)Summary
- They share 25 benchmarks with published results for both. o1 scores higher in 2 categories and Qwen3.5 397B-A17B in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3.5 397B-A17B leads 53.3 to 41.5.
- The biggest single-benchmark swing is FrontierMath (Tiers 1-3): 14.7% for o1 and 31.2% for Qwen3.5 397B-A17B.
- Qwen3.5 397B-A17B is cheaper at $0.60 / $3.60 per million input/output tokens, against $15 / $60 for o1.
- Qwen3.5 397B-A17B accepts more context: 262K tokens versus 200K.
- Qwen3.5 397B-A17B has downloadable open weights; the other is API-only.
Side by side
| o1 | Qwen3.5 397B-A17B | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 40.6 | 45.7 |
| Released | 2024-09-12 | 2026-02-01 |
| Weights | Proprietary | Open |
| Context window | 200K | 262K |
| Max output | 100K | 66K |
| Input $ / M tokens | $15 | $0.60 |
| Output $ / M tokens | $60 | $3.60 |
| Results tracked | 52 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o1 leads
o1: 45.7 (#64), Qwen3.5 397B-A17B: 41.7 (#108)
| Benchmark | o1 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Coding | 1367 | 1464 |
| Aider Polyglot | 61.7% | — |
| LMArena WebDev | — | 1400 |
| WeirdML | 47.6% | — |
| LiveBench Coding | 69.7% | — |
| CadEval | 56% | — |
| HumanEval+ | 89% | — |
| MBPP+ | 80.2% | — |
Agentic & Tool Use Qwen3.5 397B-A17B leads
o1: 24.3 (#111), Qwen3.5 397B-A17B: 33.2 (#51)
| Benchmark | o1 | Qwen3.5 397B-A17B |
|---|---|---|
| APEX-Agents | — | 24.9% |
| τ²-bench Airline | — | 81.5% |
| τ²-bench Banking | — | 9.8% |
| τ²-bench Retail | — | 84.4% |
| τ²-bench Telecom | — | 97.8% |
| Cybench | 10% | — |
| METR Time Horizons | 51.1% | — |
Reasoning Qwen3.5 397B-A17B leads
o1: 27.5 (#103), Qwen3.5 397B-A17B: 33.6 (#62)
| Benchmark | o1 | Qwen3.5 397B-A17B |
|---|---|---|
| Chess Puzzles | 15% | 13% |
| LMArena Hard Prompts | 1372 | 1448 |
| DTBench | 74.7% | 87.5% |
| LMCA | 22.3% | 37.9% |
| Epoch Capabilities Index | 141.92 | 146.64 |
| SimpleBench | 41.7% | — |
| Kagi LLM Benchmark | — | 73.7% |
| NYT Connections (extended) | — | 58.9% |
| ARC-AGI-1 | 30.7% | — |
| EnigmaEval | 5.7% | — |
| Thematic Generalization | — | 65.1% |
| LiveBench Reasoning | 91.6% | — |
| Mystery Game Puzzles | — | 18% |
| LiveBench Data Analysis | 65.5% | — |
| LiveBench | 75.7% | — |
Math Qwen3.5 397B-A17B leads
o1: 34.9 (#150), Qwen3.5 397B-A17B: 44.9 (#64)
| Benchmark | o1 | Qwen3.5 397B-A17B |
|---|---|---|
| FrontierMath (Tiers 1-3) | 14.7% | 31.2% |
| OTIS Mock AIME 2024-2025 | 73.3% | 88.9% |
| LMArena Math | 1388 | 1455 |
| LiveBench Math | 80.3% | — |
| MATH Level 5 | 94.7% | — |
| FrontierMath (Feb 2025 set) | 9.3% | — |
Knowledge Qwen3.5 397B-A17B leads
o1: 41.5 (#101), Qwen3.5 397B-A17B: 53.3 (#53)
| Benchmark | o1 | Qwen3.5 397B-A17B |
|---|---|---|
| GPQA Diamond | 76.8% | 86.4% |
| LMArena Expert | 1362 | 1461 |
| Humanity's Last Exam | 8% | — |
| SimpleQA Verified | 41.1% | — |
| Confabulations | 11.7% | — |
Multimodal Qwen3.5 397B-A17B leads
o1: 33.6 (#87), Qwen3.5 397B-A17B: 40.0 (#42)
| Benchmark | o1 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Vision | 1168 | 1262 |
| GeoBench | 80% | — |
| VPCT | 37% | — |
| SpatialViz-Bench | 41.4% | — |
Multilingual Qwen3.5 397B-A17B leads
o1: 48.8 (#137), Qwen3.5 397B-A17B: 54.0 (#56)
| Benchmark | o1 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Non-English | 1359 | 1431 |
| LMArena Chinese | 1394 | 1501 |
| LMArena French | 1343 | 1461 |
| LMArena German | 1337 | 1447 |
| LMArena Japanese | 1346 | 1427 |
| LMArena Korean | 1396 | 1382 |
| LMArena Russian | 1356 | 1429 |
| LMArena Spanish | 1345 | 1443 |
Instruction Following Too close to call
o1: 75.0 (#82), Qwen3.5 397B-A17B: 75.2 (#73)
| Benchmark | o1 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Instruction Following | 1368 | 1424 |
| LiveBench Instruction Following | 81.5% | — |
Long Context o1 leads
o1: 50.3 (#9), Qwen3.5 397B-A17B: 44.1 (#71)
| Benchmark | o1 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Longer Query | 1378 | 1442 |
| Fiction.LiveBench | 83.3% | — |
Writing & Preference Qwen3.5 397B-A17B leads
o1: 55.8 (#139), Qwen3.5 397B-A17B: 62.5 (#74)
| Benchmark | o1 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Text | 1366 | 1438 |
| LMArena Creative Writing | 1347 | 1401 |
| LMArena Multi-Turn | 1368 | 1447 |
| Short-Story Creative Writing | 70.2% | — |
| EQ-Bench Creative Writing | — | 1478 |
| LiveBench Language | 65.4% | — |
Frequently asked questions
Is o1 better than Qwen3.5 397B-A17B?
Qwen3.5 397B-A17B is the stronger model overall, scoring 45.7 to 40.6 on the Noometry Index.
Which is cheaper, o1 or Qwen3.5 397B-A17B?
Qwen3.5 397B-A17B is cheaper. It lists at $0.60 per million input tokens and $3.60 per million output tokens; o1 lists at $15 and $60.
Is o1 or Qwen3.5 397B-A17B better for coding?
o1 scores higher on coding benchmarks: 45.7 versus 41.7 in the Noometry coding category.
Which has the bigger context window?
Qwen3.5 397B-A17B does, with 262K tokens against 200K.
How many benchmarks do o1 and Qwen3.5 397B-A17B share?
25 benchmarks have published results for both models. o1 has 52 scored results on Noometry and Qwen3.5 397B-A17B has 36.