Model comparison
DeepSeek-V3.1 vs Qwen3 235B-A22B
DeepSeek-V3.1 and Qwen3 235B-A22B score almost the same on the Noometry Index (42.2 vs 43.2), so choose on price, context window or the category you care about most.
Last verified . 27 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 27 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 3 categories and Qwen3 235B-A22B in 5 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3 235B-A22B leads 48.5 to 36.1.
- The biggest single-benchmark swing is Fiction.LiveBench: 52.8% for DeepSeek-V3.1 and 75% for Qwen3 235B-A22B.
- DeepSeek-V3.1 is cheaper at $0.25 / $0.95 per million input/output tokens, against $0.70 / $2.80 for Qwen3 235B-A22B.
- DeepSeek-V3.1 accepts more context: 164K tokens versus 131K.
Side by side
| DeepSeek-V3.1 | Qwen3 235B-A22B | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 42.2 | 43.2 |
| Released | 2025-08-21 | 2025-04 |
| Weights | Open | Open |
| Context window | 164K | 131K |
| Max output | 8K | 16K |
| Input $ / M tokens | $0.25 | $0.70 |
| Output $ / M tokens | $0.95 | $2.80 |
| Results tracked | 27 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3 235B-A22B leads
DeepSeek-V3.1: 39.8 (#140), Qwen3 235B-A22B: 43.8 (#72)
| Benchmark | DeepSeek-V3.1 | Qwen3 235B-A22B |
|---|---|---|
| WeirdML | 38.4% | 41% |
| LMArena Coding | 1417 | 1445 |
| Aider Polyglot | — | 59.6% |
| SciCode | — | 42.4% |
Agentic & Tool Use Not comparable
DeepSeek-V3.1: —, Qwen3 235B-A22B: 33.7 (#48)
| Benchmark | DeepSeek-V3.1 | Qwen3 235B-A22B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 52.1% |
| Vending-Bench 2 | — | -11.34 |
Reasoning DeepSeek-V3.1 leads
DeepSeek-V3.1: 27.3 (#109), Qwen3 235B-A22B: 15.8 (#302)
| Benchmark | DeepSeek-V3.1 | Qwen3 235B-A22B |
|---|---|---|
| SimpleBench | 40% | 31% |
| Kagi LLM Benchmark | 53.2% | 69.4% |
| LMArena Hard Prompts | 1417 | 1433 |
| DTBench | 82.7% | 80.3% |
| LMCA | 24.3% | 29.3% |
| Epoch Capabilities Index | 139.92 | 143.87 |
| ForecastBench | 58 | 59.7 |
| ARC-AGI-2 | — | 1.3% |
| ARC-AGI-1 | — | 11% |
| CritPt | — | 0% |
| Chess Puzzles | — | 12% |
| Mystery Game Puzzles | — | 9% |
Math Qwen3 235B-A22B leads
DeepSeek-V3.1: 36.1 (#125), Qwen3 235B-A22B: 48.5 (#51)
| Benchmark | DeepSeek-V3.1 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Math | 1420 | 1431 |
| OTIS Mock AIME 2024-2025 | — | 86.7% |
| Omni-MATH | — | 71.8% |
| MATH Level 5 | — | 68.9% |
| FrontierMath (Feb 2025 set) | — | 8.5% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Qwen3 235B-A22B leads
DeepSeek-V3.1: 43.0 (#86), Qwen3 235B-A22B: 49.5 (#67)
| Benchmark | DeepSeek-V3.1 | Qwen3 235B-A22B |
|---|---|---|
| Vectara Hallucination Rate | 5.5% | 9.3% |
| LMArena Expert | 1403 | 1460 |
| GPQA Diamond | — | 80.1% |
| SimpleQA Verified | — | 40.4% |
| MMLU-Pro | — | 84.4% |
| Confabulations | — | 15.6% |
| GPQA (HELM) | — | 72.7% |
Multilingual Too close to call
DeepSeek-V3.1: 51.8 (#100), Qwen3 235B-A22B: 52.5 (#84)
| Benchmark | DeepSeek-V3.1 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Non-English | 1400 | 1410 |
| LMArena Chinese | 1470 | 1481 |
| LMArena French | 1447 | 1445 |
| LMArena German | 1412 | 1432 |
| LMArena Japanese | 1374 | 1400 |
| LMArena Korean | 1337 | 1391 |
| LMArena Russian | 1405 | 1411 |
| LMArena Spanish | 1429 | 1429 |
Instruction Following DeepSeek-V3.1 leads
DeepSeek-V3.1: 74.1 (#105), Qwen3 235B-A22B: 72.8 (#132)
| Benchmark | DeepSeek-V3.1 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Instruction Following | 1400 | 1408 |
| IFEval | — | 83.5% |
Long Context Qwen3 235B-A22B leads
DeepSeek-V3.1: 36.3 (#227), Qwen3 235B-A22B: 46.1 (#25)
| Benchmark | DeepSeek-V3.1 | Qwen3 235B-A22B |
|---|---|---|
| Fiction.LiveBench | 52.8% | 75% |
| LMArena Longer Query | 1422 | 1426 |
Writing & Preference Too close to call
DeepSeek-V3.1: 60.5 (#93), Qwen3 235B-A22B: 59.8 (#103)
| Benchmark | DeepSeek-V3.1 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Text | 1420 | 1419 |
| LMArena Creative Writing | 1401 | 1384 |
| EQ-Bench Creative Writing | 1436 | 1366 |
| LMArena Multi-Turn | 1408 | 1432 |
| Short-Story Creative Writing | — | 83% |
| WildBench | — | 86.6% |
Frequently asked questions
Is DeepSeek-V3.1 better than Qwen3 235B-A22B?
DeepSeek-V3.1 and Qwen3 235B-A22B score almost the same on the Noometry Index (42.2 vs 43.2), so choose on price, context window or the category you care about most.
Which is cheaper, DeepSeek-V3.1 or Qwen3 235B-A22B?
DeepSeek-V3.1 is cheaper. It lists at $0.25 per million input tokens and $0.95 per million output tokens; Qwen3 235B-A22B lists at $0.70 and $2.80.
Is DeepSeek-V3.1 or Qwen3 235B-A22B better for coding?
Qwen3 235B-A22B scores higher on coding benchmarks: 43.8 versus 39.8 in the Noometry coding category.
Which has the bigger context window?
DeepSeek-V3.1 does, with 164K tokens against 131K.
How many benchmarks do DeepSeek-V3.1 and Qwen3 235B-A22B share?
27 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and Qwen3 235B-A22B has 49.