Model comparison
Qwen2.5 72B Instruct vs QwQ-32B
QwQ-32B is the stronger model overall, scoring 39.4 to 31.5 on the Noometry Index.
Last verified . 24 shared benchmarks.
Alibaba (Qwen)Summary
- They share 24 benchmarks with published results for both. Qwen2.5 72B Instruct scores higher in 0 categories and QwQ-32B in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where QwQ-32B leads 36.2 to 18.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 8.1% for Qwen2.5 72B Instruct and 59.2% for QwQ-32B.
Side by side
| Qwen2.5 72B Instruct | QwQ-32B | |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 31.5 | 39.4 |
| Released | 2024-09 | 2024-11-28 |
| Weights | Open | Open |
| Context window | 131K | — |
| Max output | 8K | — |
| Input $ / M tokens | $1.40 | — |
| Output $ / M tokens | $5.60 | — |
| Results tracked | 43 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding QwQ-32B leads
Qwen2.5 72B Instruct: 32.8 (#256), QwQ-32B: 34.9 (#220)
| Benchmark | Qwen2.5 72B Instruct | QwQ-32B |
|---|---|---|
| BigCodeBench Instruct | 45.8% | 44.6% |
| LMArena Coding | 1293 | 1333 |
| BigCodeBench Complete | 55.9% | 54.4% |
| Aider Polyglot | — | 20.9% |
| WeirdML | 16% | — |
| LiveBench Coding | — | 72.2% |
Agentic & Tool Use Not comparable
Qwen2.5 72B Instruct: 22.0 (#124), QwQ-32B: —
| Benchmark | Qwen2.5 72B Instruct | QwQ-32B |
|---|---|---|
| TheAgentCompany | 5.7% | — |
| BALROG | 16.2% | — |
| METR Time Horizons | 35.8% | — |
Reasoning QwQ-32B leads
Qwen2.5 72B Instruct: 21.4 (#197), QwQ-32B: 22.9 (#171)
| Benchmark | Qwen2.5 72B Instruct | QwQ-32B |
|---|---|---|
| LMArena Hard Prompts | 1271 | 1326 |
| Epoch Capabilities Index | 129.01 | 137.6 |
| ForecastBench | 57.5 | 58.3 |
| Chess Puzzles | — | 5% |
| LiveBench Reasoning | — | 83.5% |
| DTBench | 62.9% | — |
| LiveBench Data Analysis | — | 65% |
| LMCA | 13.4% | — |
| BIG-Bench Hard | 79.8% | — |
| HellaSwag | 84.8% | — |
| LiveBench | — | 72% |
| PIQA | 82.6% | — |
| WinoGrande | 82.3% | — |
Math QwQ-32B leads
Qwen2.5 72B Instruct: 18.1 (#281), QwQ-32B: 36.2 (#123)
| Benchmark | Qwen2.5 72B Instruct | QwQ-32B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 8.1% | 59.2% |
| LMArena Math | 1283 | 1359 |
| Omni-MATH | 33% | — |
| LiveBench Math | — | 77.8% |
| MATH Level 5 | 63.2% | — |
Knowledge QwQ-32B leads
Qwen2.5 72B Instruct: 26.9 (#247), QwQ-32B: 37.2 (#150)
| Benchmark | Qwen2.5 72B Instruct | QwQ-32B |
|---|---|---|
| GPQA Diamond | 49.1% | 65.3% |
| Confabulations | 19.1% | 15.6% |
| LMArena Expert | 1245 | 1323 |
| MMLU-Pro | 63.1% | — |
| GPQA (HELM) | 42.6% | — |
| ARC (AI2) Challenge | 94.5% | — |
| MMLU | 85.3% | — |
| TriviaQA | 71.9% | — |
Multilingual QwQ-32B leads
Qwen2.5 72B Instruct: 41.2 (#208), QwQ-32B: 45.0 (#171)
| Benchmark | Qwen2.5 72B Instruct | QwQ-32B |
|---|---|---|
| LMArena Non-English | 1252 | 1306 |
| LMArena Chinese | 1272 | 1378 |
| LMArena French | 1279 | 1337 |
| LMArena German | 1234 | 1313 |
| LMArena Japanese | 1180 | 1263 |
| LMArena Korean | 1188 | 1278 |
| LMArena Russian | 1264 | 1297 |
| LMArena Spanish | 1255 | 1353 |
Instruction Following QwQ-32B leads
Qwen2.5 72B Instruct: 65.7 (#216), QwQ-32B: 72.8 (#131)
| Benchmark | Qwen2.5 72B Instruct | QwQ-32B |
|---|---|---|
| LMArena Instruction Following | 1254 | 1297 |
| LiveBench Instruction Following | — | 81.8% |
| IFEval | 80.6% | — |
Long Context QwQ-32B leads
Qwen2.5 72B Instruct: 38.9 (#183), QwQ-32B: 49.0 (#11)
| Benchmark | Qwen2.5 72B Instruct | QwQ-32B |
|---|---|---|
| LMArena Longer Query | 1282 | 1308 |
| Fiction.LiveBench | — | 83.3% |
Writing & Preference QwQ-32B leads
Qwen2.5 72B Instruct: 46.9 (#210), QwQ-32B: 50.9 (#175)
| Benchmark | Qwen2.5 72B Instruct | QwQ-32B |
|---|---|---|
| LMArena Text | 1269 | 1329 |
| LMArena Creative Writing | 1221 | 1288 |
| LMArena Multi-Turn | 1272 | 1314 |
| Short-Story Creative Writing | — | 80.2% |
| EQ-Bench Creative Writing | — | 1257 |
| WildBench | 80.2% | — |
| LiveBench Language | — | 51.4% |
Frequently asked questions
Is Qwen2.5 72B Instruct better than QwQ-32B?
QwQ-32B is the stronger model overall, scoring 39.4 to 31.5 on the Noometry Index.
Is Qwen2.5 72B Instruct or QwQ-32B better for coding?
QwQ-32B scores higher on coding benchmarks: 34.9 versus 32.8 in the Noometry coding category.
How many benchmarks do Qwen2.5 72B Instruct and QwQ-32B share?
24 benchmarks have published results for both models. Qwen2.5 72B Instruct has 43 scored results on Noometry and QwQ-32B has 36.