Model comparison
DeepSeek-V3.1 vs Qwen2.5-Coder-32B
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 33.4 on the Noometry Index.
Last verified . 13 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 13 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 7 categories and Qwen2.5-Coder-32B in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-V3.1 leads 60.3 to 41.6.
- DeepSeek-V3.1 is cheaper at $0.25 / $0.95 per million input/output tokens, against $0.66 / $1 for Qwen2.5-Coder-32B.
- DeepSeek-V3.1 accepts more context: 164K tokens versus 33K.
Side by side
| DeepSeek-V3.1 | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 42.8 | 33.4 |
| Released | 2025-08-21 | 2024-09-18 |
| Weights | Open | Open |
| Context window | 164K | 33K |
| Max output | 8K | 29K |
| Input $ / M tokens | $0.25 | $0.66 |
| Output $ / M tokens | $0.95 | $1 |
| Results tracked | 27 | 31 |
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), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | DeepSeek-V3.1 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Coding | 1417 | 1276 |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| WeirdML | 38.4% | — |
| BigCodeBench Instruct | — | 49% |
| LiveBench Coding | — | 56.9% |
| BigCodeBench Complete | — | 58% |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |
Reasoning DeepSeek-V3.1 leads
DeepSeek-V3.1: 27.9 (#110), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | DeepSeek-V3.1 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1417 | 1251 |
| Epoch Capabilities Index | 139.92 | 119.49 |
| SimpleBench | 40% | — |
| Kagi LLM Benchmark | 53.2% | — |
| LiveBench Reasoning | — | 42.1% |
| DTBench | 82.7% | — |
| LiveBench Data Analysis | — | 49.9% |
| LMCA | 24.3% | — |
| ForecastBench | 58 | — |
| HellaSwag | — | 83% |
| LiveBench | — | 46.2% |
| WinoGrande | — | 80.8% |
Math DeepSeek-V3.1 leads
DeepSeek-V3.1: 38.9 (#122), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | DeepSeek-V3.1 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1420 | 1251 |
| LiveBench Math | — | 46.6% |
| GSM8K | — | 93% |
Knowledge DeepSeek-V3.1 leads
DeepSeek-V3.1: 43.7 (#90), Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | DeepSeek-V3.1 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1405 | 1221 |
| Vectara Hallucination Rate | 5.5% | — |
| ARC (AI2) Challenge | — | 70.5% |
| MMLU | — | 79.1% |
Multilingual DeepSeek-V3.1 leads
DeepSeek-V3.1: 51.6 (#106), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | DeepSeek-V3.1 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1400 | 1205 |
| LMArena Chinese | 1469 | 1222 |
| LMArena Russian | 1405 | 1228 |
| LMArena French | 1447 | — |
| LMArena German | 1411 | — |
| LMArena Japanese | 1378 | — |
| LMArena Korean | 1337 | — |
| LMArena Spanish | 1431 | — |
Instruction Following DeepSeek-V3.1 leads
DeepSeek-V3.1: 73.9 (#110), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | DeepSeek-V3.1 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1400 | 1223 |
| LiveBench Instruction Following | — | 58.7% |
Long Context Qwen2.5-Coder-32B leads
DeepSeek-V3.1: 36.3 (#232), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | DeepSeek-V3.1 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1422 | 1251 |
| Fiction.LiveBench | 52.8% | — |
Writing & Preference DeepSeek-V3.1 leads
DeepSeek-V3.1: 60.3 (#98), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | DeepSeek-V3.1 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1420 | 1230 |
| LMArena Creative Writing | 1401 | 1174 |
| LMArena Multi-Turn | 1408 | 1222 |
| EQ-Bench Creative Writing | 1436 | — |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is DeepSeek-V3.1 better than Qwen2.5-Coder-32B?
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 33.4 on the Noometry Index.
Which is cheaper, DeepSeek-V3.1 or Qwen2.5-Coder-32B?
DeepSeek-V3.1 is cheaper. It lists at $0.25 per million input tokens and $0.95 per million output tokens; Qwen2.5-Coder-32B lists at $0.66 and $1.
Is DeepSeek-V3.1 or Qwen2.5-Coder-32B better for coding?
DeepSeek-V3.1 scores higher on coding benchmarks: 40.3 versus 22.6 in the Noometry coding category.
Which has the bigger context window?
DeepSeek-V3.1 does, with 164K tokens against 33K.
How many benchmarks do DeepSeek-V3.1 and Qwen2.5-Coder-32B share?
13 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and Qwen2.5-Coder-32B has 31.