DeepSeek V4.1 Flash vs Qwen3.8 Max (0902)
DeepSeek V4.1 Flash leads Qwen3.8 Max (0902) on coding, 62.5 to 62.4 on the Vibe Coding Index, and DeepSeek V4.1 Flash costs 7.1x less per output token ($0.84 vs $6.00 per million). Updated October 22, 2026; 92 benchmarks have a result for both.
DeepSeek V4.1 Flash or Qwen3.8 Max (0902): which to pick
DeepSeek V4.1 Flash and Qwen3.8 Max (0902) are level for coding, 62.5 against 62.4 on the Vibe Coding Index, so price and context decide.
Pick Qwen3.8 Max (0902) for work measured by LMArena WebDev or Terminal-Bench 4 (Vals AI), where it leads. Pick DeepSeek V4.1 Flash for output tokens 7.1x cheaper ($0.84 vs $6.00 per million) and a larger context window (1.05M against 1M tokens).
- Shared results
- 92
- DeepSeek V4.1 Flash only
- 9
- Qwen3.8 Max (0902) only
- 21
- Like-for-like categories
- 8 of 9
Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.
Which one for your work
Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.
Coding
Vibe Coding Index, six public coding benchmarks
No call
Level: 62.5 against 62.4, inside 2 points.
5 benchmarks behind each score
Intelligence
AA Intelligence Index
Qwen3.8 Max (0902)
Qwen3.8 Max (0902) leads, 45.4 against 24.7.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only Qwen3.8 Max (0902) has a AA Agentic Index result, 56.0.
Directional only
Reasoning
GPQA Diamond
No call
Only Qwen3.8 Max (0902) has a GPQA Diamond result, 93.7%.
Directional only
Knowledge
MMLU Pro
No call
Only Qwen3.8 Max (0902) has a MMLU Pro result, 88.6%.
Directional only
Web design
Design Arena: websites
DeepSeek V4.1 Flash
DeepSeek V4.1 Flash leads, 1308 against 1275.
Same board, both measured
Long documents
Listed context window
DeepSeek V4.1 Flash
DeepSeek V4.1 Flash takes 1.05M tokens against 1M.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
DeepSeek V4.1 Flash
DeepSeek V4.1 Flash $0.0006, Qwen3.8 Max (0902) $0.0050 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only Qwen3.8 Max (0902) has the number, $1.32.
Directional only
At a glance
- Coding index
- DeepSeek V4.1 Flash 62.5 from 5 benchmarks, Qwen3.8 Max (0902) 62.4 from 5 benchmarks.
- Widest category
- Agentic work, 13.9 points apart on average over 9 shared benchmarks.
- Widest single gap
- LiveBench browser: DeepSeek V4.1 Flash 84.0% against Qwen3.8 Max (0902) 32.0%, 52.0 points apart on a 0 to 100 scale.
- Output price
- DeepSeek V4.1 Flash $0.84 against Qwen3.8 Max (0902) $6.00 per million output tokens, 7.1x apart.
- Context window
- DeepSeek V4.1 Flash takes 1.05M tokens, Qwen3.8 Max (0902) 1M.
Benchmark by benchmark
First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.
| Category | DeepSeek V4.1 Flash | Bars | Qwen3.8 Max (0902) | Ties |
|---|---|---|---|---|
| Coding | 10 | 8 | ||
| Intelligence | 0 | 2 | ||
| Agentic work | 3 | 6 | ||
| Reasoning | 2 | 9 | ||
| Math | 0 | 6 | ||
| Knowledge | 1 | 6 | ||
| Web design | 7 | 1 | ||
| General | 5 | 26 | ||
| All shared boards | 28 | 64 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index108
| Benchmark | DeepSeek V4.1 Flash | Qwen3.8 Max (0902) | Leader |
|---|---|---|---|
| LiveBench coding | 90.2% | 60.3% | DeepSeek V4.1 Flash |
| LMArena WebDev | 1618 | 1674 | Qwen3.8 Max (0902) |
| AA Coding Index | Not measured | 76.2 | Only Qwen3.8 Max (0902) measured |
| LiveCodeBench (Vals AI) | Not measured | 87.8% | Only Qwen3.8 Max (0902) measured |
| SWE-bench Verified (Vals AI) | Not measured | 85.6% | Only Qwen3.8 Max (0902) measured |
| Terminal-Bench 4 (Vals AI) | 19.7% | 34.3% | Qwen3.8 Max (0902) |
| Vibe Code Bench (Vals AI) | 84.7% | 64.7% | DeepSeek V4.1 Flash |
| Code migration (Vals AI) | 45.6% | 24.0% | DeepSeek V4.1 Flash |
| ProgramBench (Vals AI) | 0.5% | 0.0% | DeepSeek V4.1 Flash |
| IOI (Vals AI) | 40.3% | 68.9% | Qwen3.8 Max (0902) |
| Terminal-Bench 2.1 (Vals AI) | 74.5% | 67.4% | DeepSeek V4.1 Flash |
| SWE-bench Multimodal (Vals AI) | 46.5% | 42.5% | DeepSeek V4.1 Flash |
| Vibe Code Bench 1-100 (Vals AI) | 16.4% | 12.8% | DeepSeek V4.1 Flash |
| SkillsBench (Vals AI) | 69.8% | 42.0% | DeepSeek V4.1 Flash |
| CyberBench (Vals AI) | 73.7% | 28.6% | DeepSeek V4.1 Flash |
| SciCode | 51.9% | 53.2% | Qwen3.8 Max (0902) |
| FrontierCode | 43.9% | Not measured | Only DeepSeek V4.1 Flash measured |
| FrontierSWE | Not measured | 17.8% | Only Qwen3.8 Max (0902) measured |
| ALE-Bench | 1092.3 | Not measured | Only DeepSeek V4.1 Flash measured |
| LMArena Text: coding | 1502 | 1502 | DeepSeek V4.1 Flash |
| LMArena Text: software and IT | 1496 | 1498 | Qwen3.8 Max (0902) |
| LMArena WebDev: React | 1628 | 1692 | Qwen3.8 Max (0902) |
| LMArena WebDev: HTML | 1612 | 1649 | Qwen3.8 Max (0902) |
| LMArena WebDev: from an image | 1606 | 1637 | Qwen3.8 Max (0902) |
| Neither model has a result on the other 15 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | DeepSeek V4.1 Flash | Qwen3.8 Max (0902) | Leader |
|---|---|---|---|
| AA Intelligence Index | 24.7 | 45.4 | Qwen3.8 Max (0902) |
| Epoch Capabilities Index | 155.9 | 157.4 | Qwen3.8 Max (0902) |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index36
| Benchmark | DeepSeek V4.1 Flash | Qwen3.8 Max (0902) | Leader |
|---|---|---|---|
| AA Agentic Index | Not measured | 56.0 | Only Qwen3.8 Max (0902) measured |
| Design Arena: agents building web apps | Not measured | 1300 | Only Qwen3.8 Max (0902) measured |
| Design Arena: agents building full-stack apps | Not measured | 1305 | Only Qwen3.8 Max (0902) measured |
| Design Arena: agents building mobile apps | Not measured | 1224 | Only Qwen3.8 Max (0902) measured |
| Design Arena: agents building Android apps | Not measured | 1196 | Only Qwen3.8 Max (0902) measured |
| Design Arena: agents building Godot games | Not measured | 1234 | Only Qwen3.8 Max (0902) measured |
| Design Arena: PowerPoint slides | Not measured | 1233 | Only Qwen3.8 Max (0902) measured |
| Design Arena: spreadsheets | Not measured | 967 | Only Qwen3.8 Max (0902) measured |
| LiveBench research | 86.8% | 69.9% | DeepSeek V4.1 Flash |
| LiveBench browser | 84.0% | 32.0% | DeepSeek V4.1 Flash |
| LiveBench long agentic | 40.1% | 42.3% | Qwen3.8 Max (0902) |
| SREBench (Vals AI) | 0.8% | Not measured | Only DeepSeek V4.1 Flash measured |
| Finance Agent (Vals AI) | 53.5% | 50.6% | DeepSeek V4.1 Flash |
| Tax Agent Bench (Vals AI) | 26.6% | 32.1% | Qwen3.8 Max (0902) |
| Legal Research Bench (Vals AI) | 41.4% | 47.6% | Qwen3.8 Max (0902) |
| Excel Modeling Benchmark (Vals AI) | 57.2% | 60.1% | Qwen3.8 Max (0902) |
| BioMysteryBench (Vals AI) | 67.8% | Not measured | Only DeepSeek V4.1 Flash measured |
| Terminal-Bench-Science (Vals AI) | 0.0% | 12.9% | Qwen3.8 Max (0902) |
| Vals Science Index | 29.6% | Not measured | Only DeepSeek V4.1 Flash measured |
| APEX-Agents | 39.5% | 63.3% | Qwen3.8 Max (0902) |
| LMArena Search | Not measured | 1261 | Only Qwen3.8 Max (0902) measured |
| Neither model has a result on the other 20 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index29
| Benchmark | DeepSeek V4.1 Flash | Qwen3.8 Max (0902) | Leader |
|---|---|---|---|
| GPQA Diamond | Not measured | 93.7% | Only Qwen3.8 Max (0902) measured |
| LiveBench reasoning | 86.7% | 88.2% | Qwen3.8 Max (0902) |
| MysteryMechanism (Vals AI) | 21.2% | 23.9% | Qwen3.8 Max (0902) |
| MMMU Pro (Vals AI) | Not measured | 88.0% | Only Qwen3.8 Max (0902) measured |
| GPQA Diamond (Artificial Analysis run) | Not measured | 92.8% | Only Qwen3.8 Max (0902) measured |
| Humanity's Last Exam (Artificial Analysis run) | 39.2% | 43.1% | Qwen3.8 Max (0902) |
| GPQA Diamond (Epoch AI run) | 89.8% | 92.7% | Qwen3.8 Max (0902) |
| ARC-AGI | 94.5% | Not measured | Only DeepSeek V4.1 Flash measured |
| ARC-AGI-2 | 72.9% | Not measured | Only DeepSeek V4.1 Flash measured |
| Chess puzzles | 48.0% | 46.0% | DeepSeek V4.1 Flash |
| Mystery game puzzles | 43.0% | 38.0% | DeepSeek V4.1 Flash |
| LMArena Text: hard prompts | 1483 | 1496 | Qwen3.8 Max (0902) |
| LMArena Text: expert | 1501 | 1507 | Qwen3.8 Max (0902) |
| LMArena Text: hard prompts in English | 1488 | 1503 | Qwen3.8 Max (0902) |
| LMArena Vision: diagrams | 1297 | 1323 | Qwen3.8 Max (0902) |
| LMArena Vision: homework | 1307 | 1336 | Qwen3.8 Max (0902) |
| Neither model has a result on the other 6 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index06
| Benchmark | DeepSeek V4.1 Flash | Qwen3.8 Max (0902) | Leader |
|---|---|---|---|
| OTIS Mock AIME | 98.3% | 100.0% | Qwen3.8 Max (0902) |
| FrontierMath, tiers 1 to 3 | 67.4% | 74.7% | Qwen3.8 Max (0902) |
| FrontierMath, tier 4 | 26.8% | 46.3% | Qwen3.8 Max (0902) |
| ProofBench | 54.0% | 58.0% | Qwen3.8 Max (0902) |
| LMArena Text: math | 1478 | 1501 | Qwen3.8 Max (0902) |
| LMArena Text: mathematical work | 1485 | 1491 | Qwen3.8 Max (0902) |
| Neither model has a result on the other 7 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index16
| Benchmark | DeepSeek V4.1 Flash | Qwen3.8 Max (0902) | Leader |
|---|---|---|---|
| MMLU Pro | Not measured | 88.6% | Only Qwen3.8 Max (0902) measured |
| LegalBench (Vals AI) | 83.3% | 83.6% | Qwen3.8 Max (0902) |
| CorpFin (Vals AI) | Not measured | 65.8% | Only Qwen3.8 Max (0902) measured |
| TaxEval (Vals AI) | Not measured | 75.5% | Only Qwen3.8 Max (0902) measured |
| MortgageTax (Vals AI) | Not measured | 64.0% | Only Qwen3.8 Max (0902) measured |
| MedCode (Vals AI) | 41.2% | 40.7% | DeepSeek V4.1 Flash |
| SimpleQA Verified | 42.9% | 47.3% | Qwen3.8 Max (0902) |
| LMArena Text: science | 1469 | 1512 | Qwen3.8 Max (0902) |
| LMArena Text: medicine and healthcare | 1484 | 1503 | Qwen3.8 Max (0902) |
| LMArena Text: legal and government | 1443 | 1492 | Qwen3.8 Max (0902) |
| LMArena Text: business and finance | 1461 | 1480 | Qwen3.8 Max (0902) |
| Neither model has a result on the other 7 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index71
| Benchmark | DeepSeek V4.1 Flash | Qwen3.8 Max (0902) | Leader |
|---|---|---|---|
| Design Arena: websites | 1308 | 1275 | DeepSeek V4.1 Flash |
| Design Arena: UI components | 1312 | 1338 | Qwen3.8 Max (0902) |
| Design Arena: game dev | 1330 | 1312 | DeepSeek V4.1 Flash |
| Design Arena: data viz | 1291 | 1281 | DeepSeek V4.1 Flash |
| Design Arena: 3D | 1355 | 1336 | DeepSeek V4.1 Flash |
| Design Arena: all code categories | 1320 | 1293 | DeepSeek V4.1 Flash |
| Design Arena: SVG | 1245 | Not measured | Only DeepSeek V4.1 Flash measured |
| Design Arena: ASCII art | 1265 | 1200 | DeepSeek V4.1 Flash |
| LiveBench design | 79.2% | 78.6% | DeepSeek V4.1 Flash |
GeneralLike-for-likePublic boards, shown as published, kept out of the index526
| Benchmark | DeepSeek V4.1 Flash | Qwen3.8 Max (0902) | Leader |
|---|---|---|---|
| LiveBench data analysis | 79.3% | 78.4% | DeepSeek V4.1 Flash |
| Public Benefits Bench (Vals AI) | 64.3% | 67.1% | Qwen3.8 Max (0902) |
| SAGE (Vals AI) | 47.9% | 51.2% | Qwen3.8 Max (0902) |
| MedScribe (Vals AI) | 85.5% | 85.0% | DeepSeek V4.1 Flash |
| Vals Multimodal Index | Not measured | 65.4% | Only Qwen3.8 Max (0902) measured |
| Long context reasoning (AA-LCR) | 84.0% | 80.3% | DeepSeek V4.1 Flash |
| DTBench | 89.9% | 92.0% | Qwen3.8 Max (0902) |
| LMCA | 47.0% | 46.2% | DeepSeek V4.1 Flash |
| Surface Evolver Bench | 46.3% | Not measured | Only DeepSeek V4.1 Flash measured |
| Furniture assembly | 34.2% | 20.0% | DeepSeek V4.1 Flash |
| LMArena Text | 1464 | 1483 | Qwen3.8 Max (0902) |
| LMArena Text: instruction following | 1473 | 1481 | Qwen3.8 Max (0902) |
| LMArena Text: creative writing | 1437 | 1477 | Qwen3.8 Max (0902) |
| LMArena Vision | 1278 | 1313 | Qwen3.8 Max (0902) |
| LMArena Text: English | 1473 | 1491 | Qwen3.8 Max (0902) |
| LMArena Text: non-English | 1449 | 1473 | Qwen3.8 Max (0902) |
| LMArena Text: Chinese | 1501 | 1539 | Qwen3.8 Max (0902) |
| LMArena Text: German | 1470 | 1478 | Qwen3.8 Max (0902) |
| LMArena Text: Japanese | 1432 | 1473 | Qwen3.8 Max (0902) |
| LMArena Text: Korean | 1436 | 1460 | Qwen3.8 Max (0902) |
| LMArena Text: French | 1465 | 1498 | Qwen3.8 Max (0902) |
| LMArena Text: Spanish | 1459 | 1489 | Qwen3.8 Max (0902) |
| LMArena Text: Russian | 1467 | 1482 | Qwen3.8 Max (0902) |
| LMArena Text: Polish | 1457 | 1471 | Qwen3.8 Max (0902) |
| LMArena Text: multi-turn | 1461 | 1489 | Qwen3.8 Max (0902) |
| LMArena Text: longer queries | 1473 | 1488 | Qwen3.8 Max (0902) |
| LMArena Text: writing and language | 1450 | 1477 | Qwen3.8 Max (0902) |
| LMArena Text: entertainment and media | 1445 | 1464 | Qwen3.8 Max (0902) |
| LMArena Vision: English | 1288 | 1311 | Qwen3.8 Max (0902) |
| LMArena Vision: Chinese | 1327 | 1358 | Qwen3.8 Max (0902) |
| LMArena Vision: reading text | 1282 | 1322 | Qwen3.8 Max (0902) |
| LMArena Vision: humor | 1260 | 1304 | Qwen3.8 Max (0902) |
| LMArena Vision: creative writing | 1275 | 1312 | Qwen3.8 Max (0902) |
| Neither model has a result on the other 13 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | DeepSeek V4.1 Flash | Qwen3.8 Max (0902) | Leader |
|---|---|---|---|
| Neither model has a result on 5 boards here | |||
What each workload costs
Three fixed token mixes at each model's listed price, so per-token rates become a bill.
Chat turn
1K input + 500 output tokens
DeepSeek V4.1 Flash costs less
Repository review
50K input + 3K output tokens
DeepSeek V4.1 Flash costs less
Agent loop
200K input + 10K output tokens
DeepSeek V4.1 Flash costs less
Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.
Specification differences
The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.
| Specification | DeepSeek V4.1 Flash | Qwen3.8 Max (0902) |
|---|---|---|
| Maker | DeepSeek | Qwen |
| Weights | Open weights | No public weights listed |
| Released | September 10, 2026 | September 3, 2026 |
| Context window | 1.05M | 1M |
| Input $/Mtok | $0.2225 | $2.00 |
| Output $/Mtok | $0.84 | $6.00 |
| Coding index | 62.5 | 62.4 |
| Independent measurements behind it | 5 | 5 |
| Cost per solved task | Not computable | $1.32 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is DeepSeek V4.1 Flash better than Qwen3.8 Max (0902) for coding?
DeepSeek V4.1 Flash and Qwen3.8 Max (0902) are level for coding, 62.5 against 62.4 on the Vibe Coding Index, so price and context decide. Pick Qwen3.8 Max (0902) for work measured by LMArena WebDev or Terminal-Bench 4 (Vals AI), where it leads. Pick DeepSeek V4.1 Flash for output tokens 7.1x cheaper ($0.84 vs $6.00 per million) and a larger context window (1.05M against 1M tokens).
Which is cheaper, DeepSeek V4.1 Flash or Qwen3.8 Max (0902)?
DeepSeek V4.1 Flash costs $0.2225 per million input tokens and $0.84 per million output tokens; Qwen3.8 Max (0902) costs $2.00 and $6.00. Cost per solved task is known for Qwen3.8 Max (0902) only.
Which has the bigger context window, DeepSeek V4.1 Flash or Qwen3.8 Max (0902)?
DeepSeek V4.1 Flash, with 1.05M tokens against 1M.
How many benchmarks is the DeepSeek V4.1 Flash vs Qwen3.8 Max (0902) comparison based on?
92 benchmarks have a result for both models as of October 22, 2026. DeepSeek V4.1 Flash's coding index rests on 5 independent benchmarks and Qwen3.8 Max (0902)'s on 5. Where only one model has a result, the page says so instead of naming a winner.