Model comparison
o1 vs Qwen1.5-72B
o1 is the stronger model overall, scoring 40.9 to 30.8 on the Noometry Index.
Last verified . 20 shared benchmarks.
Alibaba (Qwen)Summary
- They share 20 benchmarks with published results for both. o1 scores higher in 8 categories and Qwen1.5-72B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where o1 leads 41.5 to 11.5.
- The biggest single-benchmark swing is GPQA Diamond: 76.8% for o1 and 28.8% for Qwen1.5-72B.
- Qwen1.5-72B has downloadable open weights; the other is API-only.
Side by side
| o1 | Qwen1.5-72B | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 40.9 | 30.8 |
| Released | 2024-09-12 | 2024-02-04 |
| Weights | Proprietary | Open |
| Context window | 200K | — |
| Max output | 100K | — |
| Input $ / M tokens | $15 | — |
| Output $ / M tokens | $60 | — |
| Results tracked | 52 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o1 leads
o1: 46.1 (#70), Qwen1.5-72B: 31.9 (#277)
| Benchmark | o1 | Qwen1.5-72B |
|---|---|---|
| LMArena Coding | 1367 | 1165 |
| HumanEval+ | 89% | 59.1% |
| MBPP+ | 80.2% | 61.6% |
| Aider Polyglot | 61.7% | — |
| WeirdML | 47.6% | — |
| BigCodeBench Instruct | — | 33.2% |
| LiveBench Coding | 69.7% | — |
| BigCodeBench Complete | — | 40.3% |
| CadEval | 56% | — |
Agentic & Tool Use Not comparable
o1: 24.6 (#117), Qwen1.5-72B: —
| Benchmark | o1 | Qwen1.5-72B |
|---|---|---|
| Cybench | 10% | — |
| METR Time Horizons | 51.1% | — |
Reasoning o1 leads
o1: 27.9 (#111), Qwen1.5-72B: 22.2 (#203)
| Benchmark | o1 | Qwen1.5-72B |
|---|---|---|
| LMArena Hard Prompts | 1371 | 1148 |
| SimpleBench | 41.7% | — |
| ARC-AGI-1 | 30.7% | — |
| Chess Puzzles | 15% | — |
| EnigmaEval | 5.7% | — |
| LiveBench Reasoning | 91.6% | — |
| DTBench | 74.7% | — |
| LiveBench Data Analysis | 65.5% | — |
| LMCA | 22.3% | — |
| Epoch Capabilities Index | 141.91 | — |
| LiveBench | 75.7% | — |
Math o1 leads
o1: 36.1 (#175), Qwen1.5-72B: 33.2 (#205)
| Benchmark | o1 | Qwen1.5-72B |
|---|---|---|
| LMArena Math | 1388 | 1164 |
| FrontierMath (Tiers 1-3) | 14.7% | — |
| OTIS Mock AIME 2024-2025 | 73.3% | — |
| LiveBench Math | 80.3% | — |
| MATH Level 5 | 94.7% | — |
| FrontierMath (Feb 2025 set) | 9.3% | — |
Knowledge o1 leads
o1: 41.5 (#110), Qwen1.5-72B: 11.5 (#300)
| Benchmark | o1 | Qwen1.5-72B |
|---|---|---|
| GPQA Diamond | 76.8% | 28.8% |
| LMArena Expert | 1361 | 1136 |
| Humanity's Last Exam | 8% | — |
| SimpleQA Verified | 41.1% | — |
| Confabulations | 11.7% | — |
Multimodal Not comparable
o1: 34.2 (#93), Qwen1.5-72B: —
| Benchmark | o1 | Qwen1.5-72B |
|---|---|---|
| LMArena Vision | 1168 | — |
| GeoBench | 80% | — |
| VPCT | 37% | — |
| SpatialViz-Bench | 41.4% | — |
Multilingual o1 leads
o1: 48.6 (#142), Qwen1.5-72B: 33.2 (#253)
| Benchmark | o1 | Qwen1.5-72B |
|---|---|---|
| LMArena Non-English | 1358 | 1135 |
| LMArena Chinese | 1394 | 1186 |
| LMArena French | 1344 | 1159 |
| LMArena German | 1337 | 1084 |
| LMArena Japanese | 1346 | 1061 |
| LMArena Korean | 1396 | 1050 |
| LMArena Russian | 1356 | 1104 |
| LMArena Spanish | 1345 | 1110 |
Instruction Following o1 leads
o1: 74.8 (#86), Qwen1.5-72B: 59.3 (#256)
| Benchmark | o1 | Qwen1.5-72B |
|---|---|---|
| LMArena Instruction Following | 1367 | 1141 |
| LiveBench Instruction Following | 81.5% | — |
Long Context o1 leads
o1: 50.3 (#9), Qwen1.5-72B: 35.1 (#243)
| Benchmark | o1 | Qwen1.5-72B |
|---|---|---|
| LMArena Longer Query | 1378 | 1157 |
| Fiction.LiveBench | 83.3% | — |
Writing & Preference o1 leads
o1: 55.6 (#144), Qwen1.5-72B: 37.3 (#258)
| Benchmark | o1 | Qwen1.5-72B |
|---|---|---|
| LMArena Text | 1366 | 1166 |
| LMArena Creative Writing | 1348 | 1137 |
| LMArena Multi-Turn | 1369 | 1160 |
| Short-Story Creative Writing | 70.2% | — |
| LiveBench Language | 65.4% | — |
Frequently asked questions
Is o1 better than Qwen1.5-72B?
o1 is the stronger model overall, scoring 40.9 to 30.8 on the Noometry Index.
Is o1 or Qwen1.5-72B better for coding?
o1 scores higher on coding benchmarks: 46.1 versus 31.9 in the Noometry coding category.
How many benchmarks do o1 and Qwen1.5-72B share?
20 benchmarks have published results for both models. o1 has 52 scored results on Noometry and Qwen1.5-72B has 22.