Model comparison
DeepSeek-V3.2-Exp vs Qwen3.8 Max
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 44.3 on the Noometry Index. DeepSeek-V3.2-Exp costs 10× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.
Last verified . 29 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 29 benchmarks with published results for both. DeepSeek-V3.2-Exp scores higher in 1 category and Qwen3.8 Max in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.8 Max leads 54.4 to 22.1.
- The biggest single-benchmark swing is NYT Connections (extended): 36.7% for DeepSeek-V3.2-Exp and 88.3% for Qwen3.8 Max.
- DeepSeek-V3.2-Exp is cheaper at $0.26 / $0.38 per million input/output tokens, against $2 / $6 for Qwen3.8 Max.
- Qwen3.8 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 | Qwen3.8 Max | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 44.3 | 56.8 |
| Released | 2025-09-29 | 2026-08-02 |
| Weights | Open | Proprietary |
| Context window | 164K | 1M |
| Max output | 66K | 131K |
| Input $ / M tokens | $0.26 | $2 |
| Output $ / M tokens | $0.38 | $6 |
| Results tracked | 49 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 Max leads
DeepSeek-V3.2-Exp: 46.5 (#65), Qwen3.8 Max: 53.5 (#29)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.8 Max |
|---|---|---|
| LMArena WebDev | 1362 | 1674 |
| SciCode | 38.9% | 53.2% |
| LMArena Coding | 1454 | 1502 |
| DeepSWE | — | 57.5% |
| SWE-bench Verified (bash only) | 70% | — |
| Aider Polyglot | 74.2% | — |
| SWE-bench Multilingual | 59% | — |
| FrontierSWE | — | 17.8% |
| WeirdML | 39.5% | — |
Agentic & Tool Use Qwen3.8 Max leads
DeepSeek-V3.2-Exp: 32.7 (#59), Qwen3.8 Max: 45.4 (#14)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.8 Max |
|---|---|---|
| APEX-Agents | 21.3% | 63.3% |
| Terminal-Bench | 39.6% | — |
| Berkeley Function Calling Leaderboard | 56.7% | — |
| TheAgentCompany | 42.9% | — |
| τ²-bench Banking | — | 55.1% |
| GDP.pdf | — | 23.2% |
| Vending-Bench 2 | 1,034 | — |
Reasoning Qwen3.8 Max leads
DeepSeek-V3.2-Exp: 22.1 (#208), Qwen3.8 Max: 54.4 (#26)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.8 Max |
|---|---|---|
| NYT Connections (extended) | 36.7% | 88.3% |
| CritPt | 2.9% | 20% |
| Chess Puzzles | 14% | 40% |
| LMArena Hard Prompts | 1434 | 1496 |
| DTBench | 87.7% | 92% |
| LMCA | 29.1% | 46.2% |
| Epoch Capabilities Index | 146.27 | 156.41 |
| ARC-AGI-2 | 4% | — |
| Kagi LLM Benchmark | 52.2% | — |
| ARC-AGI-1 | 57% | — |
| Thematic Generalization | 65% | — |
| Mystery Game Puzzles | — | 38% |
Math Qwen3.8 Max leads
DeepSeek-V3.2-Exp: 41.7 (#87), Qwen3.8 Max: 73.2 (#20)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.8 Max |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 87.8% | 100% |
| ProofBench | 8% | 58% |
| LMArena Math | 1435 | 1499 |
| FrontierMath (Tiers 1-3) | — | 74.7% |
| FrontierMath Tier 4 | — | 46.3% |
| MathArena Final-Answer Competitions | 57.7% | — |
| FrontierMath (Feb 2025 set) | 22.1% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge Qwen3.8 Max leads
DeepSeek-V3.2-Exp: 51.7 (#66), Qwen3.8 Max: 61.7 (#27)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.8 Max |
|---|---|---|
| GPQA Diamond | 83.4% | 92.7% |
| LMArena Expert | 1436 | 1507 |
| SimpleQA Verified | — | 47.3% |
| Vectara Hallucination Rate | 5.3% | — |
Multimodal Not comparable
DeepSeek-V3.2-Exp: —, Qwen3.8 Max: 37.2 (#75)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | — | 1314 |
| Furniture Assembly | — | 20% |
Multilingual Qwen3.8 Max leads
DeepSeek-V3.2-Exp: 52.2 (#90), Qwen3.8 Max: 56.7 (#18)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1409 | 1472 |
| LMArena Chinese | 1461 | 1538 |
| LMArena French | 1433 | 1503 |
| LMArena German | 1440 | 1483 |
| LMArena Japanese | 1374 | 1467 |
| LMArena Korean | 1371 | 1461 |
| LMArena Russian | 1424 | 1481 |
| LMArena Spanish | 1440 | 1492 |
Instruction Following Qwen3.8 Max leads
DeepSeek-V3.2-Exp: 74.5 (#93), Qwen3.8 Max: 77.6 (#17)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1413 | 1479 |
Long Context DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 47.6 (#16), Qwen3.8 Max: 45.6 (#31)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1428 | 1489 |
| Fiction.LiveBench | 83.3% | — |
| CL-bench | 13.2% | — |
| CL-bench Life | 9.5% | — |
Writing & Preference Qwen3.8 Max leads
DeepSeek-V3.2-Exp: 62.4 (#77), Qwen3.8 Max: 67.1 (#30)
| Benchmark | DeepSeek-V3.2-Exp | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1425 | 1483 |
| LMArena Creative Writing | 1403 | 1479 |
| LMArena Multi-Turn | 1427 | 1489 |
| EQ-Bench Creative Writing | 1515 | — |
Frequently asked questions
Is DeepSeek-V3.2-Exp better than Qwen3.8 Max?
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 44.3 on the Noometry Index. DeepSeek-V3.2-Exp costs 10× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.
Which is cheaper, DeepSeek-V3.2-Exp or Qwen3.8 Max?
DeepSeek-V3.2-Exp is cheaper. It lists at $0.26 per million input tokens and $0.38 per million output tokens; Qwen3.8 Max lists at $2 and $6.
Is DeepSeek-V3.2-Exp or Qwen3.8 Max better for coding?
Qwen3.8 Max scores higher on coding benchmarks: 53.5 versus 46.5 in the Noometry coding category.
Which has the bigger context window?
Qwen3.8 Max does, with 1M tokens against 164K.
How many benchmarks do DeepSeek-V3.2-Exp and Qwen3.8 Max share?
29 benchmarks have published results for both models. DeepSeek-V3.2-Exp has 49 scored results on Noometry and Qwen3.8 Max has 39.