Model comparison
DeepSeek-V3.2-Exp vs Qwen3.7 Max
Qwen3.7 Max is the stronger model overall, scoring 51.4 to 44.9 on the Noometry Index. DeepSeek-V3.2-Exp costs 13× less per token, which makes it the better buy when Qwen3.7 Max's lead doesn't matter for your workload.
Last verified . 23 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 23 benchmarks with published results for both. DeepSeek-V3.2-Exp scores higher in 2 categories and Qwen3.7 Max in 7 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.7 Max leads 48.4 to 22.0.
- The biggest single-benchmark swing is NYT Connections (extended): 36.7% for DeepSeek-V3.2-Exp and 85.1% for Qwen3.7 Max.
- DeepSeek-V3.2-Exp is cheaper at $0.26 / $0.38 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-V3.2-Exp has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V3.2-Exp | Qwen3.7 Max | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 44.9 | 51.4 |
| Released | 2025-09-29 | 2026-05-19 |
| Weights | Open | Proprietary |
| Context window | 164K | 1M |
| Max output | 66K | 131K |
| Input $ / M tokens | $0.26 | $2.50 |
| Output $ / M tokens | $0.38 | $7.50 |
| Results tracked | 48 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.7 Max leads
DeepSeek-V3.2-Exp: 46.4 (#61), Qwen3.7 Max: 50.2 (#43)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.7 Max |
|---|---|---|
| LMArena WebDev | 1362 | 1515 |
| SciCode | 38.9% | 48.8% |
| LMArena Coding | 1454 | 1498 |
| SWE-bench Verified | — | 77.3% |
| SWE-bench Verified (bash only) | 70% | — |
| Aider Polyglot | 74.2% | — |
| SWE-bench Multilingual | 59% | — |
| WeirdML | 39.5% | — |
| ALE-Bench | — | 1,189 |
Agentic & Tool Use DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 37.9 (#31), Qwen3.7 Max: 21.8 (#129)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.7 Max |
|---|---|---|
| Terminal-Bench | 39.6% | — |
| Berkeley Function Calling Leaderboard | 56.7% | — |
| TheAgentCompany | 42.9% | — |
| GBAEval | — | 0.4% |
| Vending-Bench 2 | 1,034 | — |
Reasoning Qwen3.7 Max leads
DeepSeek-V3.2-Exp: 22.0 (#185), Qwen3.7 Max: 48.4 (#36)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.7 Max |
|---|---|---|
| NYT Connections (extended) | 36.7% | 85.1% |
| CritPt | 2.9% | 13.4% |
| Chess Puzzles | 14% | 19% |
| LMArena Hard Prompts | 1433 | 1483 |
| DTBench | 87.7% | 92.3% |
| LMCA | 29.1% | 44% |
| Epoch Capabilities Index | 146.28 | 153.74 |
| ARC-AGI-2 | 4% | — |
| SimpleBench | — | 70.4% |
| Kagi LLM Benchmark | 52.2% | — |
| ARC-AGI-1 | 57% | — |
| Thematic Generalization | 65% | — |
| EBR-Bench | — | 9.5% |
| Mystery Game Puzzles | — | 32% |
Math Qwen3.7 Max leads
DeepSeek-V3.2-Exp: 40.8 (#75), Qwen3.7 Max: 62.3 (#26)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.7 Max |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 87.8% | 95.6% |
| ProofBench | 8% | 26% |
| LMArena Math | 1435 | 1488 |
| FrontierMath (Tiers 1-3) | — | 64.6% |
| FrontierMath Tier 4 | — | 34.1% |
| MathArena Final-Answer Competitions | 57.7% | — |
| FrontierMath (Feb 2025 set) | 22.1% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge Qwen3.7 Max leads
DeepSeek-V3.2-Exp: 51.6 (#61), Qwen3.7 Max: 61.7 (#24)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.7 Max |
|---|---|---|
| GPQA Diamond | 83.4% | 90.9% |
| LMArena Expert | 1436 | 1488 |
| SimpleQA Verified | — | 55.8% |
| Vectara Hallucination Rate | 5.3% | — |
Multilingual Qwen3.7 Max leads
DeepSeek-V3.2-Exp: 52.4 (#85), Qwen3.7 Max: 57.1 (#13)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.7 Max |
|---|---|---|
| LMArena Non-English | 1409 | 1475 |
| LMArena Chinese | 1461 | 1533 |
| LMArena Russian | 1423 | 1482 |
| LMArena French | 1435 | — |
| LMArena German | 1440 | — |
| LMArena Japanese | 1377 | — |
| LMArena Korean | 1372 | — |
| LMArena Spanish | 1441 | — |
Instruction Following Qwen3.7 Max leads
DeepSeek-V3.2-Exp: 74.7 (#89), Qwen3.7 Max: 76.9 (#34)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.7 Max |
|---|---|---|
| LMArena Instruction Following | 1413 | 1460 |
Long Context DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 47.6 (#15), Qwen3.7 Max: 45.4 (#34)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.7 Max |
|---|---|---|
| LMArena Longer Query | 1428 | 1484 |
| Fiction.LiveBench | 83.3% | — |
| CL-bench | 13.2% | — |
| CL-bench Life | 9.5% | — |
Writing & Preference Qwen3.7 Max leads
DeepSeek-V3.2-Exp: 62.5 (#73), Qwen3.7 Max: 65.2 (#52)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.7 Max |
|---|---|---|
| LMArena Text | 1425 | 1476 |
| LMArena Creative Writing | 1403 | 1450 |
| LMArena Multi-Turn | 1427 | 1482 |
| EQ-Bench Creative Writing | 1515 | — |
| EQ-Bench 4 | — | 1110 |
Frequently asked questions
Is DeepSeek-V3.2-Exp better than Qwen3.7 Max?
Qwen3.7 Max is the stronger model overall, scoring 51.4 to 44.9 on the Noometry Index. DeepSeek-V3.2-Exp costs 13× 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-V3.2-Exp or Qwen3.7 Max?
DeepSeek-V3.2-Exp is cheaper. It lists at $0.26 per million input tokens and $0.38 per million output tokens; Qwen3.7 Max lists at $2.50 and $7.50.
Is DeepSeek-V3.2-Exp or Qwen3.7 Max better for coding?
Qwen3.7 Max scores higher on coding benchmarks: 50.2 versus 46.4 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-V3.2-Exp and Qwen3.7 Max share?
23 benchmarks have published results for both models. DeepSeek-V3.2-Exp has 48 scored results on Noometry and Qwen3.7 Max has 33.