Model comparison
DeepSeek-V3.1 vs Qwen3.7 Max
Qwen3.7 Max is the stronger model overall, scoring 51.4 to 42.2 on the Noometry Index. DeepSeek-V3.1 costs 8.8× less per token, which makes it the better buy when Qwen3.7 Max's lead doesn't matter for your workload.
Last verified . 16 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 16 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 0 categories and Qwen3.7 Max in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.7 Max leads 62.3 to 36.1.
- The biggest single-benchmark swing is SimpleBench: 40% for DeepSeek-V3.1 and 70.4% for Qwen3.7 Max.
- DeepSeek-V3.1 is cheaper at $0.25 / $0.95 per million input/output tokens, against $2.50 / $7.50 for Qwen3.7 Max.
- Qwen3.7 Max accepts more context: 1M tokens versus 164K.
- DeepSeek-V3.1 has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V3.1 | Qwen3.7 Max | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 42.2 | 51.4 |
| Released | 2025-08-21 | 2026-05-19 |
| Weights | Open | Proprietary |
| Context window | 164K | 1M |
| Max output | 8K | 131K |
| Input $ / M tokens | $0.25 | $2.50 |
| Output $ / M tokens | $0.95 | $7.50 |
| Results tracked | 27 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.7 Max leads
DeepSeek-V3.1: 39.8 (#140), Qwen3.7 Max: 50.2 (#43)
| Benchmark | DeepSeek-V3.1 | Qwen3.7 Max |
|---|---|---|
| LMArena Coding | 1417 | 1498 |
| SWE-bench Verified | — | 77.3% |
| LMArena WebDev | — | 1515 |
| SciCode | — | 48.8% |
| WeirdML | 38.4% | — |
| ALE-Bench | — | 1,189 |
Agentic & Tool Use Not comparable
DeepSeek-V3.1: —, Qwen3.7 Max: 21.8 (#129)
| Benchmark | DeepSeek-V3.1 | Qwen3.7 Max |
|---|---|---|
| GBAEval | — | 0.4% |
Reasoning Qwen3.7 Max leads
DeepSeek-V3.1: 27.3 (#109), Qwen3.7 Max: 48.4 (#36)
| Benchmark | DeepSeek-V3.1 | Qwen3.7 Max |
|---|---|---|
| SimpleBench | 40% | 70.4% |
| LMArena Hard Prompts | 1417 | 1483 |
| DTBench | 82.7% | 92.3% |
| LMCA | 24.3% | 44% |
| Epoch Capabilities Index | 139.92 | 153.74 |
| Kagi LLM Benchmark | 53.2% | — |
| NYT Connections (extended) | — | 85.1% |
| CritPt | — | 13.4% |
| Chess Puzzles | — | 19% |
| EBR-Bench | — | 9.5% |
| Mystery Game Puzzles | — | 32% |
| ForecastBench | 58 | — |
Math Qwen3.7 Max leads
DeepSeek-V3.1: 36.1 (#125), Qwen3.7 Max: 62.3 (#26)
| Benchmark | DeepSeek-V3.1 | Qwen3.7 Max |
|---|---|---|
| LMArena Math | 1420 | 1488 |
| FrontierMath (Tiers 1-3) | — | 64.6% |
| FrontierMath Tier 4 | — | 34.1% |
| OTIS Mock AIME 2024-2025 | — | 95.6% |
| ProofBench | — | 26% |
Knowledge Qwen3.7 Max leads
DeepSeek-V3.1: 43.0 (#86), Qwen3.7 Max: 61.7 (#24)
| Benchmark | DeepSeek-V3.1 | Qwen3.7 Max |
|---|---|---|
| LMArena Expert | 1403 | 1488 |
| GPQA Diamond | — | 90.9% |
| SimpleQA Verified | — | 55.8% |
| Vectara Hallucination Rate | 5.5% | — |
Multilingual Qwen3.7 Max leads
DeepSeek-V3.1: 51.8 (#100), Qwen3.7 Max: 57.1 (#13)
| Benchmark | DeepSeek-V3.1 | Qwen3.7 Max |
|---|---|---|
| LMArena Non-English | 1400 | 1475 |
| LMArena Chinese | 1470 | 1533 |
| LMArena Russian | 1405 | 1482 |
| LMArena French | 1447 | — |
| LMArena German | 1412 | — |
| LMArena Japanese | 1374 | — |
| LMArena Korean | 1337 | — |
| LMArena Spanish | 1429 | — |
Instruction Following Qwen3.7 Max leads
DeepSeek-V3.1: 74.1 (#105), Qwen3.7 Max: 76.9 (#34)
| Benchmark | DeepSeek-V3.1 | Qwen3.7 Max |
|---|---|---|
| LMArena Instruction Following | 1400 | 1460 |
Long Context Qwen3.7 Max leads
DeepSeek-V3.1: 36.3 (#227), Qwen3.7 Max: 45.4 (#34)
| Benchmark | DeepSeek-V3.1 | Qwen3.7 Max |
|---|---|---|
| LMArena Longer Query | 1422 | 1484 |
| Fiction.LiveBench | 52.8% | — |
Writing & Preference Qwen3.7 Max leads
DeepSeek-V3.1: 60.5 (#93), Qwen3.7 Max: 65.2 (#52)
| Benchmark | DeepSeek-V3.1 | Qwen3.7 Max |
|---|---|---|
| LMArena Text | 1420 | 1476 |
| LMArena Creative Writing | 1401 | 1450 |
| LMArena Multi-Turn | 1408 | 1482 |
| EQ-Bench Creative Writing | 1436 | — |
| EQ-Bench 4 | — | 1110 |
Frequently asked questions
Is DeepSeek-V3.1 better than Qwen3.7 Max?
Qwen3.7 Max is the stronger model overall, scoring 51.4 to 42.2 on the Noometry Index. DeepSeek-V3.1 costs 8.8× less per token, which makes it the better buy when Qwen3.7 Max's lead doesn't matter for your workload.
Which is cheaper, DeepSeek-V3.1 or Qwen3.7 Max?
DeepSeek-V3.1 is cheaper. It lists at $0.25 per million input tokens and $0.95 per million output tokens; Qwen3.7 Max lists at $2.50 and $7.50.
Is DeepSeek-V3.1 or Qwen3.7 Max better for coding?
Qwen3.7 Max scores higher on coding benchmarks: 50.2 versus 39.8 in the Noometry coding category.
Which has the bigger context window?
Qwen3.7 Max does, with 1M tokens against 164K.
How many benchmarks do DeepSeek-V3.1 and Qwen3.7 Max share?
16 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and Qwen3.7 Max has 33.