DeepSeek V4.1 Flash vs Gemini 3.5 Flash
DeepSeek V4.1 Flash leads Gemini 3.5 Flash on coding, 62.5 to 39.8 on the Vibe Coding Index, and DeepSeek V4.1 Flash costs 10.7x less per output token ($0.84 vs $9.00 per million). Updated October 22, 2026; 84 benchmarks have a result for both.
vs boards won of 84 shared, 1 tied
DeepSeek V4.1 Flash or Gemini 3.5 Flash: which to pick
Pick DeepSeek V4.1 Flash for coding: it scores 62.5 on the Vibe Coding Index against 39.8 for Gemini 3.5 Flash.
Pick Gemini 3.5 Flash for work measured by SciCode or AA Intelligence Index, where it leads.
Model ADeepSeek V4.1 FlashDeepSeek62.5/1005 independent benchmarksNo cost per solved taskModel BGemini 3.5 FlashGoogle39.8/1004 independent benchmarks$1.21 per solved task
- Shared results
- 84
- DeepSeek V4.1 Flash only
- 17
- Gemini 3.5 Flash only
- 28
- 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
DeepSeek V4.1 Flash
DeepSeek V4.1 Flash scores 62.5 against 39.8.
5 and 4 benchmarks behind the scores
Intelligence
AA Intelligence Index
Gemini 3.5 Flash
Gemini 3.5 Flash leads, 33.6 against 24.7.
Same board, both measured
Reasoning
GPQA Diamond
No call
Only Gemini 3.5 Flash has a GPQA Diamond result, 92.7%.
Directional only
Knowledge
MMLU Pro
No call
Only Gemini 3.5 Flash has a MMLU Pro result, 89.5%.
Directional only
Web design
Design Arena: websites
DeepSeek V4.1 Flash
DeepSeek V4.1 Flash leads, 1308 against 1267.
Same board, both measured
Long documents
Listed context window
No call
Both take 1.05M tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
DeepSeek V4.1 Flash
DeepSeek V4.1 Flash $0.0006, Gemini 3.5 Flash $0.0060 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only Gemini 3.5 Flash has the number, $1.21.
Directional only
At a glance
- Coding index
- DeepSeek V4.1 Flash 62.5 from 5 benchmarks, Gemini 3.5 Flash 39.8 from 4 benchmarks.
- Widest category
- Coding, 11.7 points apart on average over 14 shared benchmarks.
- Widest single gap
- Vibe Code Bench (Vals AI): DeepSeek V4.1 Flash 84.7% against Gemini 3.5 Flash 48.7%, 36.1 points apart on a 0 to 100 scale.
- Output price
- DeepSeek V4.1 Flash $0.84 against Gemini 3.5 Flash $9.00 per million output tokens, 10.7x apart.
- Context window
- Both take 1.05M tokens.
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 | Gemini 3.5 Flash | Ties |
|---|---|---|---|---|
| Coding | 13 | 1 | ||
| Intelligence | 1 | 1 | ||
| Agentic work | 3 | 3 | ||
| Reasoning | 6 | 5 | ||
| Math | 3 | 2 | 1 tie | |
| Knowledge | 1 | 6 | ||
| Web design | 7 | 1 | ||
| General | 4 | 26 | ||
| All shared boards | 38 | 45 | 1 tie |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index131
| Benchmark | DeepSeek V4.1 Flash | Gemini 3.5 Flash | Leader |
|---|---|---|---|
| LiveBench coding | 90.2% | Not measured | Only DeepSeek V4.1 Flash measured |
| LMArena WebDev | 1618 | 1499 | DeepSeek V4.1 Flash |
| LiveCodeBench (Vals AI) | Not measured | 87.6% | Only Gemini 3.5 Flash measured |
| SWE-bench Verified (Vals AI) | Not measured | 78.8% | Only Gemini 3.5 Flash measured |
| Terminal-Bench 4 (Vals AI) | 19.7% | 6.1% | DeepSeek V4.1 Flash |
| Vibe Code Bench (Vals AI) | 84.7% | 48.7% | DeepSeek V4.1 Flash |
| Code migration (Vals AI) | 45.6% | 26.8% | DeepSeek V4.1 Flash |
| ProgramBench (Vals AI) | 0.5% | 0.0% | DeepSeek V4.1 Flash |
| IOI (Vals AI) | 40.3% | Not measured | Only DeepSeek V4.1 Flash measured |
| Terminal-Bench 2.1 (Vals AI) | 74.5% | 74.2% | DeepSeek V4.1 Flash |
| SWE-bench Multimodal (Vals AI) | 46.5% | Not measured | Only DeepSeek V4.1 Flash measured |
| Vibe Code Bench 1-100 (Vals AI) | 16.4% | Not measured | Only DeepSeek V4.1 Flash measured |
| SkillsBench (Vals AI) | 69.8% | 52.7% | DeepSeek V4.1 Flash |
| CyberBench (Vals AI) | 73.7% | Not measured | Only DeepSeek V4.1 Flash measured |
| SciCode | 51.9% | 53.9% | Gemini 3.5 Flash |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 46.2% | Only Gemini 3.5 Flash measured |
| FrontierCode | 43.9% | Not measured | Only DeepSeek V4.1 Flash measured |
| WeirdML | Not measured | 62.6% | Only Gemini 3.5 Flash measured |
| ALE-Bench | 1092.3 | 911.0 | DeepSeek V4.1 Flash |
| LMArena Text: coding | 1502 | 1493 | DeepSeek V4.1 Flash |
| LMArena Text: software and IT | 1496 | 1494 | DeepSeek V4.1 Flash |
| LMArena WebDev: React | 1628 | 1495 | DeepSeek V4.1 Flash |
| LMArena WebDev: HTML | 1612 | 1522 | DeepSeek V4.1 Flash |
| LMArena WebDev: from an image | 1606 | 1504 | DeepSeek V4.1 Flash |
| SciCode | Not measured | 12.3% | Only Gemini 3.5 Flash measured |
| Neither model has a result on the other 14 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index11
| Benchmark | DeepSeek V4.1 Flash | Gemini 3.5 Flash | Leader |
|---|---|---|---|
| AA Intelligence Index | 24.7 | 33.6 | Gemini 3.5 Flash |
| Epoch Capabilities Index | 155.9 | 154.9 | DeepSeek V4.1 Flash |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index33
| Benchmark | DeepSeek V4.1 Flash | Gemini 3.5 Flash | Leader |
|---|---|---|---|
| Design Arena: agents building web apps | Not measured | 1202 | Only Gemini 3.5 Flash measured |
| Design Arena: agents building full-stack apps | Not measured | 1197 | Only Gemini 3.5 Flash measured |
| Design Arena: agents building mobile apps | Not measured | 1174 | Only Gemini 3.5 Flash measured |
| Design Arena: agents building Android apps | Not measured | 1185 | Only Gemini 3.5 Flash measured |
| Design Arena: agents building Godot games | Not measured | 1098 | Only Gemini 3.5 Flash measured |
| Design Arena: agents building games | Not measured | 1180 | Only Gemini 3.5 Flash measured |
| Design Arena: HTML slides | Not measured | 1093 | Only Gemini 3.5 Flash measured |
| Design Arena: PowerPoint slides | Not measured | 1154 | Only Gemini 3.5 Flash measured |
| LiveBench research | 86.8% | Not measured | Only DeepSeek V4.1 Flash measured |
| LiveBench browser | 84.0% | Not measured | Only DeepSeek V4.1 Flash measured |
| LiveBench long agentic | 40.1% | Not measured | Only DeepSeek V4.1 Flash measured |
| SREBench (Vals AI) | 0.8% | Not measured | Only DeepSeek V4.1 Flash measured |
| Finance Agent (Vals AI) | 53.5% | 57.9% | Gemini 3.5 Flash |
| Tax Agent Bench (Vals AI) | 26.6% | 21.6% | DeepSeek V4.1 Flash |
| Legal Research Bench (Vals AI) | 41.4% | 30.8% | DeepSeek V4.1 Flash |
| Excel Modeling Benchmark (Vals AI) | 57.2% | 63.5% | Gemini 3.5 Flash |
| BioMysteryBench (Vals AI) | 67.8% | Not measured | Only DeepSeek V4.1 Flash measured |
| Terminal-Bench-Science (Vals AI) | 0.0% | 5.7% | Gemini 3.5 Flash |
| Vals Science Index | 29.6% | Not measured | Only DeepSeek V4.1 Flash measured |
| τ²-Bench Telecom | Not measured | 95.6% | Only Gemini 3.5 Flash measured |
| APEX-Agents | 39.5% | 27.5% | DeepSeek V4.1 Flash |
| Vending-Bench 2 | Not measured | $5,396 | Only Gemini 3.5 Flash measured |
| Neither model has a result on the other 19 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index65
| Benchmark | DeepSeek V4.1 Flash | Gemini 3.5 Flash | Leader |
|---|---|---|---|
| GPQA Diamond | Not measured | 92.7% | Only Gemini 3.5 Flash measured |
| LiveBench reasoning | 86.7% | Not measured | Only DeepSeek V4.1 Flash measured |
| MysteryMechanism (Vals AI) | 21.2% | Not measured | Only DeepSeek V4.1 Flash measured |
| MMMU Pro (Vals AI) | Not measured | 88.3% | Only Gemini 3.5 Flash measured |
| GPQA Diamond (Artificial Analysis run) | Not measured | 92.2% | Only Gemini 3.5 Flash measured |
| Humanity's Last Exam (Artificial Analysis run) | 39.2% | 42.7% | Gemini 3.5 Flash |
| GPQA Diamond (Epoch AI run) | 89.8% | 92.8% | Gemini 3.5 Flash |
| ARC-AGI | 94.5% | 92.5% | DeepSeek V4.1 Flash |
| ARC-AGI-2 | 72.9% | 72.1% | DeepSeek V4.1 Flash |
| SimpleBench | Not measured | 76.7% | Only Gemini 3.5 Flash measured |
| Chess puzzles | 48.0% | 45.0% | DeepSeek V4.1 Flash |
| Mystery game puzzles | 43.0% | 32.0% | DeepSeek V4.1 Flash |
| EBR-bench | Not measured | 4.8% | Only Gemini 3.5 Flash measured |
| LMArena Text: hard prompts | 1483 | 1488 | Gemini 3.5 Flash |
| LMArena Text: expert | 1501 | 1495 | DeepSeek V4.1 Flash |
| LMArena Text: hard prompts in English | 1488 | 1483 | DeepSeek V4.1 Flash |
| LMArena Vision: diagrams | 1297 | 1323 | Gemini 3.5 Flash |
| LMArena Vision: homework | 1307 | 1330 | Gemini 3.5 Flash |
| Neither model has a result on the other 4 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index32
| Benchmark | DeepSeek V4.1 Flash | Gemini 3.5 Flash | Leader |
|---|---|---|---|
| OTIS Mock AIME | 98.3% | 95.6% | DeepSeek V4.1 Flash |
| FrontierMath, tiers 1 to 3 | 67.4% | 62.8% | DeepSeek V4.1 Flash |
| FrontierMath, tier 4 | 26.8% | 26.8% | Tie |
| ProofBench | 54.0% | 31.0% | DeepSeek V4.1 Flash |
| LMArena Text: math | 1478 | 1504 | Gemini 3.5 Flash |
| LMArena Text: mathematical work | 1485 | 1495 | Gemini 3.5 Flash |
| 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 | Gemini 3.5 Flash | Leader |
|---|---|---|---|
| MMLU Pro | Not measured | 89.5% | Only Gemini 3.5 Flash measured |
| LegalBench (Vals AI) | 83.3% | 83.6% | Gemini 3.5 Flash |
| CorpFin (Vals AI) | Not measured | 64.7% | Only Gemini 3.5 Flash measured |
| TaxEval (Vals AI) | Not measured | 74.4% | Only Gemini 3.5 Flash measured |
| MortgageTax (Vals AI) | Not measured | 68.1% | Only Gemini 3.5 Flash measured |
| MedCode (Vals AI) | 41.2% | 55.8% | Gemini 3.5 Flash |
| SimpleQA Verified | 42.9% | 66.2% | Gemini 3.5 Flash |
| LMArena Text: science | 1469 | 1495 | Gemini 3.5 Flash |
| LMArena Text: medicine and healthcare | 1484 | 1483 | DeepSeek V4.1 Flash |
| LMArena Text: legal and government | 1443 | 1488 | Gemini 3.5 Flash |
| LMArena Text: business and finance | 1461 | 1467 | Gemini 3.5 Flash |
| FACTS Benchmark Suite | Not measured | 63.1% | Only Gemini 3.5 Flash measured |
| Neither model has a result on the other 6 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index71
| Benchmark | DeepSeek V4.1 Flash | Gemini 3.5 Flash | Leader |
|---|---|---|---|
| Design Arena: websites | 1308 | 1267 | DeepSeek V4.1 Flash |
| Design Arena: UI components | 1312 | 1269 | DeepSeek V4.1 Flash |
| Design Arena: game dev | 1330 | 1263 | DeepSeek V4.1 Flash |
| Design Arena: data viz | 1291 | 1236 | DeepSeek V4.1 Flash |
| Design Arena: 3D | 1355 | 1243 | DeepSeek V4.1 Flash |
| Design Arena: all code categories | 1320 | 1266 | DeepSeek V4.1 Flash |
| Design Arena: SVG | 1245 | 1253 | Gemini 3.5 Flash |
| Design Arena: ASCII art | 1265 | 1254 | DeepSeek V4.1 Flash |
| LiveBench design | 79.2% | Not measured | Only DeepSeek V4.1 Flash measured |
GeneralLike-for-likePublic boards, shown as published, kept out of the index426
| Benchmark | DeepSeek V4.1 Flash | Gemini 3.5 Flash | Leader |
|---|---|---|---|
| LiveBench data analysis | 79.3% | Not measured | Only DeepSeek V4.1 Flash measured |
| Public Benefits Bench (Vals AI) | 64.3% | 59.5% | DeepSeek V4.1 Flash |
| SAGE (Vals AI) | 47.9% | 49.9% | Gemini 3.5 Flash |
| MedScribe (Vals AI) | 85.5% | 76.6% | DeepSeek V4.1 Flash |
| Vals Multimodal Index | Not measured | 62.9% | Only Gemini 3.5 Flash measured |
| IFBench | Not measured | 76.3% | Only Gemini 3.5 Flash measured |
| Long context reasoning (AA-LCR) | 84.0% | 74.3% | DeepSeek V4.1 Flash |
| DTBench | 89.9% | 94.7% | Gemini 3.5 Flash |
| LMCA | 47.0% | 47.1% | Gemini 3.5 Flash |
| Surface Evolver Bench | 46.3% | 58.1% | Gemini 3.5 Flash |
| Furniture assembly | 34.2% | Not measured | Only DeepSeek V4.1 Flash measured |
| LMArena Text | 1464 | 1482 | Gemini 3.5 Flash |
| LMArena Text: instruction following | 1473 | 1467 | DeepSeek V4.1 Flash |
| LMArena Text: creative writing | 1437 | 1470 | Gemini 3.5 Flash |
| LMArena Vision | 1278 | 1308 | Gemini 3.5 Flash |
| LMArena Text: English | 1473 | 1481 | Gemini 3.5 Flash |
| LMArena Text: non-English | 1449 | 1475 | Gemini 3.5 Flash |
| LMArena Text: Chinese | 1501 | 1527 | Gemini 3.5 Flash |
| LMArena Text: German | 1470 | 1494 | Gemini 3.5 Flash |
| LMArena Text: Japanese | 1432 | 1486 | Gemini 3.5 Flash |
| LMArena Text: Korean | 1436 | 1448 | Gemini 3.5 Flash |
| LMArena Text: French | 1465 | 1490 | Gemini 3.5 Flash |
| LMArena Text: Spanish | 1459 | 1480 | Gemini 3.5 Flash |
| LMArena Text: Russian | 1467 | 1493 | Gemini 3.5 Flash |
| LMArena Text: Polish | 1457 | 1516 | Gemini 3.5 Flash |
| LMArena Text: multi-turn | 1461 | 1481 | Gemini 3.5 Flash |
| LMArena Text: longer queries | 1473 | 1483 | Gemini 3.5 Flash |
| LMArena Text: writing and language | 1450 | 1478 | Gemini 3.5 Flash |
| LMArena Text: entertainment and media | 1445 | 1459 | Gemini 3.5 Flash |
| LMArena Vision: English | 1288 | 1305 | Gemini 3.5 Flash |
| LMArena Vision: Chinese | 1327 | 1377 | Gemini 3.5 Flash |
| LMArena Vision: reading text | 1282 | 1317 | Gemini 3.5 Flash |
| LMArena Vision: humor | 1260 | 1321 | Gemini 3.5 Flash |
| LMArena Vision: creative writing | 1275 | 1301 | Gemini 3.5 Flash |
| LMArena Document | Not measured | 1463 | Only Gemini 3.5 Flash measured |
| Neither model has a result on the other 11 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | DeepSeek V4.1 Flash | Gemini 3.5 Flash | 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$0.0006Fits in one request
Gemini 3.5 Flash$0.0060Fits in one request
DeepSeek V4.1 Flash costs less
Repository review
50K input + 3K output tokens
DeepSeek V4.1 Flash$0.014Fits in one request
Gemini 3.5 Flash$0.102Fits in one request
DeepSeek V4.1 Flash costs less
Agent loop
200K input + 10K output tokens
DeepSeek V4.1 Flash$0.053Fits in one request
Gemini 3.5 Flash$0.390Fits in one request
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 | Google Gemini 3.5 Flash |
|---|---|---|
| Maker | DeepSeek | |
| Weights | Open weights | No public weights listed |
| Released | September 10, 2026 | May 19, 2026 |
| Context window | 1.05M | 1.05M |
| Input $/Mtok | $0.2225 | $1.50 |
| Output $/Mtok | $0.84 | $9.00 |
| Coding index | 62.5 | 39.8 |
| Independent measurements behind it | 5 | 4 |
| Cost per solved task | Not computable | $1.21 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is DeepSeek V4.1 Flash better than Gemini 3.5 Flash for coding?
Pick DeepSeek V4.1 Flash for coding: it scores 62.5 on the Vibe Coding Index against 39.8 for Gemini 3.5 Flash. Pick Gemini 3.5 Flash for work measured by SciCode or AA Intelligence Index, where it leads.
Which is cheaper, DeepSeek V4.1 Flash or Gemini 3.5 Flash?
DeepSeek V4.1 Flash costs $0.2225 per million input tokens and $0.84 per million output tokens; Gemini 3.5 Flash costs $1.50 and $9.00. Cost per solved task is known for Gemini 3.5 Flash only.
Which has the bigger context window, DeepSeek V4.1 Flash or Gemini 3.5 Flash?
Neither: both take 1.05M tokens.
How many benchmarks is the DeepSeek V4.1 Flash vs Gemini 3.5 Flash comparison based on?
84 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 Gemini 3.5 Flash's on 4. Where only one model has a result, the page says so instead of naming a winner.