Model comparison
DeepSeek-V3.1 vs QwQ-32B
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 39.8 on the Noometry Index.
Last verified . 21 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 21 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 7 categories and QwQ-32B in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in long context, where QwQ-32B leads 49.0 to 36.3.
- The biggest single-benchmark swing is Fiction.LiveBench: 52.8% for DeepSeek-V3.1 and 83.3% for QwQ-32B.
Side by side
| DeepSeek-V3.1 | QwQ-32B | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 42.8 | 39.8 |
| Released | 2025-08-21 | 2024-11-28 |
| Weights | Open | Open |
| Context window | 164K | — |
| Max output | 8K | — |
| Input $ / M tokens | $0.25 | — |
| Output $ / M tokens | $0.95 | — |
| Results tracked | 27 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.1 leads
DeepSeek-V3.1: 40.3 (#144), QwQ-32B: 35.4 (#226)
| Benchmark | DeepSeek-V3.1 | QwQ-32B |
|---|---|---|
| LMArena Coding | 1417 | 1333 |
| Aider Polyglot | — | 20.9% |
| WeirdML | 38.4% | — |
| BigCodeBench Instruct | — | 44.6% |
| LiveBench Coding | — | 72.2% |
| BigCodeBench Complete | — | 54.4% |
Reasoning DeepSeek-V3.1 leads
DeepSeek-V3.1: 27.9 (#110), QwQ-32B: 23.7 (#174)
| Benchmark | DeepSeek-V3.1 | QwQ-32B |
|---|---|---|
| LMArena Hard Prompts | 1417 | 1325 |
| Epoch Capabilities Index | 139.92 | 137.6 |
| ForecastBench | 58 | 58.3 |
| SimpleBench | 40% | — |
| Kagi LLM Benchmark | 53.2% | — |
| Chess Puzzles | — | 5% |
| LiveBench Reasoning | — | 83.5% |
| DTBench | 82.7% | — |
| LiveBench Data Analysis | — | 65% |
| LMCA | 24.3% | — |
| LiveBench | — | 72% |
Math Too close to call
DeepSeek-V3.1: 38.9 (#122), QwQ-32B: 38.0 (#143)
| Benchmark | DeepSeek-V3.1 | QwQ-32B |
|---|---|---|
| LMArena Math | 1420 | 1359 |
| OTIS Mock AIME 2024-2025 | — | 59.2% |
| LiveBench Math | — | 77.8% |
Knowledge DeepSeek-V3.1 leads
DeepSeek-V3.1: 43.7 (#90), QwQ-32B: 37.2 (#158)
| Benchmark | DeepSeek-V3.1 | QwQ-32B |
|---|---|---|
| LMArena Expert | 1405 | 1324 |
| GPQA Diamond | — | 65.3% |
| Confabulations | — | 15.6% |
| Vectara Hallucination Rate | 5.5% | — |
Multilingual DeepSeek-V3.1 leads
DeepSeek-V3.1: 51.6 (#106), QwQ-32B: 44.8 (#176)
| Benchmark | DeepSeek-V3.1 | QwQ-32B |
|---|---|---|
| LMArena Non-English | 1400 | 1305 |
| LMArena Chinese | 1469 | 1378 |
| LMArena French | 1447 | 1336 |
| LMArena German | 1411 | 1313 |
| LMArena Japanese | 1378 | 1262 |
| LMArena Korean | 1337 | 1279 |
| LMArena Russian | 1405 | 1297 |
| LMArena Spanish | 1431 | 1354 |
Instruction Following DeepSeek-V3.1 leads
DeepSeek-V3.1: 73.9 (#110), QwQ-32B: 72.6 (#137)
| Benchmark | DeepSeek-V3.1 | QwQ-32B |
|---|---|---|
| LMArena Instruction Following | 1400 | 1297 |
| LiveBench Instruction Following | — | 81.8% |
Long Context QwQ-32B leads
DeepSeek-V3.1: 36.3 (#232), QwQ-32B: 49.0 (#11)
| Benchmark | DeepSeek-V3.1 | QwQ-32B |
|---|---|---|
| Fiction.LiveBench | 52.8% | 83.3% |
| LMArena Longer Query | 1422 | 1308 |
Writing & Preference DeepSeek-V3.1 leads
DeepSeek-V3.1: 60.3 (#98), QwQ-32B: 50.6 (#180)
| Benchmark | DeepSeek-V3.1 | QwQ-32B |
|---|---|---|
| LMArena Text | 1420 | 1329 |
| LMArena Creative Writing | 1401 | 1288 |
| EQ-Bench Creative Writing | 1436 | 1257 |
| LMArena Multi-Turn | 1408 | 1314 |
| Short-Story Creative Writing | — | 80.2% |
| LiveBench Language | — | 51.4% |
Frequently asked questions
Is DeepSeek-V3.1 better than QwQ-32B?
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 39.8 on the Noometry Index.
Is DeepSeek-V3.1 or QwQ-32B better for coding?
DeepSeek-V3.1 scores higher on coding benchmarks: 40.3 versus 35.4 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.1 and QwQ-32B share?
21 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and QwQ-32B has 36.