DeepSeek V4.1 Flash vs Qwen3.8 Max
DeepSeek V4.1 Flash 
Qwen3.8 Max Verdict
- Qwen3.8 Max (max) leads on quality: 64.2 vs 61.8.
- DeepSeek V4.1 Flash (max) is stronger in composite, long context.
- Qwen3.8 Max (max) is stronger in agents & tools, coding, human preference, knowledge, maths, multimodal, reasoning.
- DeepSeek V4.1 Flash (max) is 5.7× cheaper ($0.525 vs $3.00 per 1M blended).
- DeepSeek V4.1 Flash (max) streams 5.9× faster (217 vs 37 tokens per second).
| Metric | DeepSeek V4.1 Flash (max) | Qwen3.8 Max (max) |
|---|---|---|
| BenchLeader Index | 61.8 | 64.2 |
| Agents & tools score | 55.5 | 58.2 |
| Coding score | 64.0 | 64.6 |
| Composite score | 71.2 | 70.7 |
| Human preference score | 66.9 | 68.0 |
| Knowledge score | 59.3 | 62.7 |
| Long context score | 67.3 | 65.4 |
| Maths score | 61.9 | 65.1 |
| Multimodal score | 60.9 | 66.2 |
| Reasoning score | 65.2 | 72.2 |
| Blended price $/M | $0.525 | $3.00 |
| Output speed | 217 tok/s | 37 tok/s |
| Time to first answer | 10.4 s | 58.9 s |
| Context window | 1M | 1M |
| GPQA Diamond | 89.8% | – |
| FrontierMath Tiers 1–3 | 67.4% | – |
| FrontierMath Tier 4 | 26.8% | – |
| OTIS Mock AIME | 98.3% | – |
| Terminal-Bench | – | 27.0% |
| SciCode | 51.9% | 53.2% |
| APEX-Agents | – | 63.3% |
| ProofBench | – | 58.0% |
| Epoch Capabilities Index | – | 156.4 |
| LMArena Text | 1475 | 1483 |
| LMArena Hard Prompts | 1499 | 1504 |
| LMArena Coding | 1528 | 1524 |
| LMArena WebDev | 1619 | 1672 |
| LMArena Vision | 1277 | 1314 |
| LMArena Agent | 3.8 | 2.3 |
| LiveBench | 81.1% | 78.5% |
| LiveBench Reasoning | 86.7% | 88.2% |
| LiveBench Coding | 80.0% | 72.9% |
| LiveBench Agentic Coding | 77.3% | 64.7% |
| LiveBench Mathematics | 93.3% | 91.3% |
| LiveBench Data Analysis | 79.3% | 78.4% |
| LiveBench Language | 81.2% | 79.7% |
| LiveBench Instruction Following | 70.0% | 74.1% |
| AA Intelligence Index v4.3.2 | 39.5 | 45.4 |
| AA-LCR | 84.0% | 80.3% |
| MMMU-Pro | 77.0% | 82.8% |
| AA-Omniscience | -5.3 | 12.0 |
| GPQA Diamond (AA) | – | 92.8% |
| Humanity's Last Exam (AA) | 39.3% | 43.1% |
| SciCode (AA) | 51.9% | 53.2% |
| LiveCodeBench | – | 87.8% |
| MMLU-Pro | – | 88.6% |
| IOI | – | 68.9% |
| LegalBench | – | 83.6% |
| CorpFin | – | 65.8% |
| TaxEval | – | 75.5% |
| Terminal-Bench 2.1 (Vals) | – | 67.4% |
| SWE-bench (Vals) | – | 85.6% |
| GPQA Diamond (Vals) | – | 93.7% |
| Vals Index | – | 48.3 |
| ARC-AGI-1 | 94.5% | – |
| ARC-AGI-2 | 72.9% | – |
| CritPt | 14.3% | 20.0% |
| GDPval-AA v2.1 | 55.0% | 58.6% |
| τ³-Banking (AA) | – | 51.3% |
| ITBench SRE (AA) | 46.9% | 40.3% |
| Analyst Agent (AA) | – | 45.0% |
| APEX-Agents (AA) | – | 42.4% |
| Code Migration | – | 24.0% |
| CyberBench | – | 28.6% |
| Excel Modeling Benchmark | – | 60.1% |
| Finance Agent v2 | – | 50.6% |
| Harvey's Legal Agent Benchmark | 6.7% | 10.4% |
| Legal Research Bench | – | 47.6% |
| MedCode | – | 40.7% |
| MedScribe | – | 85.0% |
| MMMU-Pro (Vals) | – | 88.0% |
| MortgageTax | – | 64.0% |
| MysteryMechanism | – | 23.9% |
| ProgramBench | – | 0.0% |
| Public Benefits Bench | 64.3% | 67.1% |
| SAGE | – | 51.3% |
| SkillsBench | – | 42.0% |
| SREBench | 0.8% | – |
| Tax Agent Bench | – | 66.0% |
| Terminal-Bench 4.0 (Vals) | – | 34.3% |
| Terminal-Bench Science | – | 1.4% |
| Vals Multimodal Index | – | 65.4% |
| Vibe Code Bench 1-100 | – | 12.8% |
| Vibe Code Bench v1.1 | – | 64.7% |
| LMArena Maths | 1493 | 1497 |
| LMArena Creative Writing | 1439 | 1470 |
| LMArena Instruction Following | 1477 | 1474 |
| LMArena Multi-turn | 1466 | 1492 |
| LMArena Longer Queries | 1482 | 1492 |
| Mystery Game Puzzles | 43.0% | – |
| ALE-Bench | 1092.3 | – |
| GDP.pdf | 19.8% | – |
| Terminal-Bench 4.0 (AA) | 26.8% | 38.9% |
| Terminal-Bench 2.1 (AA) | – | 88.8% |
| AutomationBench | 68.9% | 56.2% |
| GDP.pdf | 12.8% | 22.8% |
| MLCR | 22.8% | 20.0% |
| Harvey LAB | 1.7% | – |
| EnterpriseOps-Gym | – | 47.6% |
| AA-Omniscience: accuracy | 46.4% | 31.9% |
| AA-Omniscience: non-hallucination | 3.5% | 71.2% |
| AA-Briefcase v1.1 | 1422 | 1617 |
| AA Openness Index | 44.4 | – |
Data as of 2026-10-11. Best configuration of each model; every score links to its source on the model pages.
DeepSeek V4.1 Flash vs Qwen3.8 Max: questions
- Is DeepSeek V4.1 Flash better than Qwen3.8 Max?
- Qwen3.8 Max (max) leads on quality: 64.2 vs 61.8. The BenchLeader Index combines every independent quality benchmark; Qwen3.8 Max (max) is ahead overall as of 2026-10-11, but check the category scores for your use.
- Is DeepSeek V4.1 Flash better than Qwen3.8 Max for coding?
- Qwen3.8 Max scores higher in coding (65 vs 64 on the category index, where 50 is average).
- Is DeepSeek V4.1 Flash better than Qwen3.8 Max for agentic tasks?
- Qwen3.8 Max scores higher in agentic tasks (58 vs 56 on the category index, where 50 is average).
- Which is cheaper, DeepSeek V4.1 Flash or Qwen3.8 Max?
- DeepSeek V4.1 Flash is cheaper: $0.525 against $3.00 per million tokens, blended at three input tokens per output token.
- Which is faster, DeepSeek V4.1 Flash or Qwen3.8 Max?
- DeepSeek V4.1 Flash streams faster: 217 against 37 output tokens per second.
- Which has the larger context window?
- Both accept 1M tokens of context.