DeepSeek V4.1 Flash vs Gemini 4 Argon
Gemini 4 Argon leads DeepSeek V4.1 Flash on coding, 81.0 to 62.5 on the Vibe Coding Index, and DeepSeek V4.1 Flash costs 11.9x less per output token ($0.84 vs $10.00 per million). Updated October 22, 2026; 83 benchmarks have a result for both.
DeepSeek V4.1 Flash or Gemini 4 Argon: which to pick
Pick Gemini 4 Argon for coding: it scores 81.0 on the Vibe Coding Index against 62.5 for DeepSeek V4.1 Flash.
Pick DeepSeek V4.1 Flash for output tokens 11.9x cheaper ($0.84 vs $10.00 per million) and work measured by LiveBench coding or Terminal-Bench 2.1 (Vals AI), where it leads.
- Shared results
- 83
- DeepSeek V4.1 Flash only
- 18
- Gemini 4 Argon only
- 11
- 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
Gemini 4 Argon
Gemini 4 Argon scores 81.0 against 62.5.
5 benchmarks behind each score
Intelligence
AA Intelligence Index
Gemini 4 Argon
Gemini 4 Argon leads, 52.6 against 24.7.
Same board, both measured
Web design
Design Arena: websites
Gemini 4 Argon
Gemini 4 Argon leads, 1315 against 1308.
Same board, both measured
General
Vals Index
No call
Only Gemini 4 Argon has a Vals Index result, 69.2%.
Directional only
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 4 Argon $0.0070 per turn.
List prices
At a glance
- Coding index
- DeepSeek V4.1 Flash 62.5 from 5 benchmarks, Gemini 4 Argon 81.0 from 5 benchmarks.
- Widest category
- Math, 23.1 points apart on average over 6 shared benchmarks.
- Widest single gap
- IOI (Vals AI): DeepSeek V4.1 Flash 40.3% against Gemini 4 Argon 100.0%, 59.7 points apart on a 0 to 100 scale.
- Output price
- DeepSeek V4.1 Flash $0.84 against Gemini 4 Argon $10.00 per million output tokens, 11.9x 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 4 Argon | Ties |
|---|---|---|---|---|
| Coding | 2 | 16 | ||
| Intelligence | 0 | 2 | ||
| Agentic work | 1 | 11 | ||
| Reasoning | 0 | 11 | ||
| Math | 0 | 6 | ||
| Knowledge | 0 | 7 | ||
| Web design | 1 | 6 | ||
| General | 1 | 19 | ||
| All shared boards | 5 | 78 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index216
| Benchmark | DeepSeek V4.1 Flash | Gemini 4 Argon | Leader |
|---|---|---|---|
| LiveBench coding | 90.2% | 89.7% | DeepSeek V4.1 Flash |
| LMArena WebDev | 1618 | 1678 | Gemini 4 Argon |
| Terminal-Bench 4 (Vals AI) | 19.7% | 57.6% | Gemini 4 Argon |
| Vibe Code Bench (Vals AI) | 84.7% | 91.9% | Gemini 4 Argon |
| Code migration (Vals AI) | 45.6% | 68.2% | Gemini 4 Argon |
| ProgramBench (Vals AI) | 0.5% | 2.5% | Gemini 4 Argon |
| IOI (Vals AI) | 40.3% | 100.0% | Gemini 4 Argon |
| Terminal-Bench 2.1 (Vals AI) | 74.5% | 57.7% | DeepSeek V4.1 Flash |
| SWE-bench Multimodal (Vals AI) | 46.5% | 56.0% | Gemini 4 Argon |
| Vibe Code Bench 1-100 (Vals AI) | 16.4% | 21.2% | Gemini 4 Argon |
| SkillsBench (Vals AI) | 69.8% | Not measured | Only DeepSeek V4.1 Flash measured |
| CyberBench (Vals AI) | 73.7% | 77.9% | Gemini 4 Argon |
| SciCode | 51.9% | 61.8% | Gemini 4 Argon |
| FrontierCode | 43.9% | 47.5% | Gemini 4 Argon |
| FrontierSWE | Not measured | 55.0% | Only Gemini 4 Argon measured |
| MirrorCode | Not measured | 58.9% | Only Gemini 4 Argon measured |
| ALE-Bench | 1092.3 | 2100.1 | Gemini 4 Argon |
| LMArena Text: coding | 1502 | 1546 | Gemini 4 Argon |
| LMArena Text: software and IT | 1496 | 1556 | Gemini 4 Argon |
| LMArena WebDev: React | 1628 | 1692 | Gemini 4 Argon |
| LMArena WebDev: HTML | 1612 | 1643 | Gemini 4 Argon |
| LMArena WebDev: from an image | 1606 | Not measured | Only DeepSeek V4.1 Flash measured |
| SciCode | Not measured | 12.3% | Only Gemini 4 Argon measured |
| Neither model has a result on the other 16 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | DeepSeek V4.1 Flash | Gemini 4 Argon | Leader |
|---|---|---|---|
| AA Intelligence Index | 24.7 | 52.6 | Gemini 4 Argon |
| Epoch Capabilities Index | 155.9 | 166.4 | Gemini 4 Argon |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index111
| Benchmark | DeepSeek V4.1 Flash | Gemini 4 Argon | Leader |
|---|---|---|---|
| Design Arena: spreadsheets | Not measured | 1157 | Only Gemini 4 Argon measured |
| LiveBench research | 86.8% | 83.1% | DeepSeek V4.1 Flash |
| LiveBench browser | 84.0% | 92.0% | Gemini 4 Argon |
| LiveBench long agentic | 40.1% | 69.9% | Gemini 4 Argon |
| SREBench (Vals AI) | 0.8% | 44.3% | Gemini 4 Argon |
| Finance Agent (Vals AI) | 53.5% | 65.4% | Gemini 4 Argon |
| Tax Agent Bench (Vals AI) | 26.6% | 44.6% | Gemini 4 Argon |
| Legal Research Bench (Vals AI) | 41.4% | 54.8% | Gemini 4 Argon |
| Harvey's Legal Agent Benchmark (Vals AI) | Not measured | 27.1% | Only Gemini 4 Argon measured |
| Excel Modeling Benchmark (Vals AI) | 57.2% | 75.2% | Gemini 4 Argon |
| BioMysteryBench (Vals AI) | 67.8% | 76.3% | Gemini 4 Argon |
| Terminal-Bench-Science (Vals AI) | 0.0% | 44.3% | Gemini 4 Argon |
| Vals Science Index | 29.6% | 55.4% | Gemini 4 Argon |
| Time Horizon Index (Vals AI) | Not measured | 52.5% | Only Gemini 4 Argon measured |
| CUA-bench (Vals AI) | Not measured | 4.8% | Only Gemini 4 Argon measured |
| Robo-bench (Vals AI) | Not measured | 0.8% | Only Gemini 4 Argon measured |
| APEX-Agents | 39.5% | 82.4% | Gemini 4 Argon |
| Vending-Bench 2 | Not measured | $13,718 | Only Gemini 4 Argon measured |
| Neither model has a result on the other 23 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index011
| Benchmark | DeepSeek V4.1 Flash | Gemini 4 Argon | Leader |
|---|---|---|---|
| LiveBench reasoning | 86.7% | 95.1% | Gemini 4 Argon |
| MysteryMechanism (Vals AI) | 21.2% | 45.5% | Gemini 4 Argon |
| Humanity's Last Exam (Artificial Analysis run) | 39.2% | 57.1% | Gemini 4 Argon |
| GPQA Diamond (Epoch AI run) | 89.8% | 96.4% | Gemini 4 Argon |
| ARC-AGI | 94.5% | 98.5% | Gemini 4 Argon |
| ARC-AGI-2 | 72.9% | 91.3% | Gemini 4 Argon |
| Chess puzzles | 48.0% | 59.0% | Gemini 4 Argon |
| Mystery game puzzles | 43.0% | 64.0% | Gemini 4 Argon |
| EBR-bench | Not measured | 64.8% | Only Gemini 4 Argon measured |
| LMArena Text: hard prompts | 1483 | 1546 | Gemini 4 Argon |
| LMArena Text: expert | 1501 | 1550 | Gemini 4 Argon |
| LMArena Text: hard prompts in English | 1488 | 1548 | Gemini 4 Argon |
| LMArena Vision: diagrams | 1297 | Not measured | Only DeepSeek V4.1 Flash measured |
| LMArena Vision: homework | 1307 | Not measured | Only DeepSeek V4.1 Flash measured |
| Neither model has a result on the other 8 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index06
| Benchmark | DeepSeek V4.1 Flash | Gemini 4 Argon | Leader |
|---|---|---|---|
| OTIS Mock AIME | 98.3% | 100.0% | Gemini 4 Argon |
| FrontierMath, tiers 1 to 3 | 67.4% | 86.0% | Gemini 4 Argon |
| FrontierMath, tier 4 | 26.8% | 78.0% | Gemini 4 Argon |
| ProofBench | 54.0% | 99.0% | Gemini 4 Argon |
| LMArena Text: math | 1478 | 1536 | Gemini 4 Argon |
| LMArena Text: mathematical work | 1485 | 1557 | Gemini 4 Argon |
| Neither model has a result on the other 7 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index07
| Benchmark | DeepSeek V4.1 Flash | Gemini 4 Argon | Leader |
|---|---|---|---|
| LegalBench (Vals AI) | 83.3% | 88.3% | Gemini 4 Argon |
| MedCode (Vals AI) | 41.2% | 58.8% | Gemini 4 Argon |
| SimpleQA Verified | 42.9% | 76.0% | Gemini 4 Argon |
| LMArena Text: science | 1469 | 1548 | Gemini 4 Argon |
| LMArena Text: medicine and healthcare | 1484 | 1555 | Gemini 4 Argon |
| LMArena Text: legal and government | 1443 | 1548 | Gemini 4 Argon |
| LMArena Text: business and finance | 1461 | 1525 | Gemini 4 Argon |
| Neither model has a result on the other 11 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index16
| Benchmark | DeepSeek V4.1 Flash | Gemini 4 Argon | Leader |
|---|---|---|---|
| Design Arena: websites | 1308 | 1315 | Gemini 4 Argon |
| Design Arena: UI components | 1312 | 1333 | Gemini 4 Argon |
| Design Arena: game dev | 1330 | 1377 | Gemini 4 Argon |
| Design Arena: data viz | 1291 | 1332 | Gemini 4 Argon |
| Design Arena: 3D | 1355 | 1388 | Gemini 4 Argon |
| Design Arena: all code categories | 1320 | 1329 | Gemini 4 Argon |
| Design Arena: SVG | 1245 | Not measured | Only DeepSeek V4.1 Flash measured |
| Design Arena: ASCII art | 1265 | Not measured | Only DeepSeek V4.1 Flash measured |
| LiveBench design | 79.2% | 73.7% | DeepSeek V4.1 Flash |
GeneralLike-for-likePublic boards, shown as published, kept out of the index119
| Benchmark | DeepSeek V4.1 Flash | Gemini 4 Argon | Leader |
|---|---|---|---|
| Vals Index | Not measured | 69.2% | Only Gemini 4 Argon measured |
| LiveBench data analysis | 79.3% | 85.2% | Gemini 4 Argon |
| Public Benefits Bench (Vals AI) | 64.3% | 69.8% | Gemini 4 Argon |
| SAGE (Vals AI) | 47.9% | 53.6% | Gemini 4 Argon |
| MedScribe (Vals AI) | 85.5% | 87.4% | Gemini 4 Argon |
| Long context reasoning (AA-LCR) | 84.0% | 79.7% | DeepSeek V4.1 Flash |
| DTBench | 89.9% | 97.6% | Gemini 4 Argon |
| LMCA | 47.0% | 63.7% | Gemini 4 Argon |
| Surface Evolver Bench | 46.3% | Not measured | Only DeepSeek V4.1 Flash measured |
| Furniture assembly | 34.2% | 72.5% | Gemini 4 Argon |
| LMArena Text | 1464 | 1532 | Gemini 4 Argon |
| LMArena Text: instruction following | 1473 | 1534 | Gemini 4 Argon |
| LMArena Text: creative writing | 1437 | 1527 | Gemini 4 Argon |
| LMArena Vision | 1278 | Not measured | Only DeepSeek V4.1 Flash measured |
| LMArena Text: English | 1473 | 1538 | Gemini 4 Argon |
| LMArena Text: non-English | 1449 | 1522 | Gemini 4 Argon |
| LMArena Text: Chinese | 1501 | 1611 | Gemini 4 Argon |
| LMArena Text: German | 1470 | Not measured | Only DeepSeek V4.1 Flash measured |
| LMArena Text: Japanese | 1432 | Not measured | Only DeepSeek V4.1 Flash measured |
| LMArena Text: Korean | 1436 | Not measured | Only DeepSeek V4.1 Flash measured |
| LMArena Text: French | 1465 | Not measured | Only DeepSeek V4.1 Flash measured |
| LMArena Text: Spanish | 1459 | 1503 | Gemini 4 Argon |
| LMArena Text: Russian | 1467 | 1534 | Gemini 4 Argon |
| LMArena Text: Polish | 1457 | Not measured | Only DeepSeek V4.1 Flash measured |
| LMArena Text: multi-turn | 1461 | 1556 | Gemini 4 Argon |
| LMArena Text: longer queries | 1473 | 1547 | Gemini 4 Argon |
| LMArena Text: writing and language | 1450 | 1529 | Gemini 4 Argon |
| LMArena Text: entertainment and media | 1445 | 1527 | Gemini 4 Argon |
| LMArena Vision: English | 1288 | Not measured | Only DeepSeek V4.1 Flash measured |
| LMArena Vision: Chinese | 1327 | Not measured | Only DeepSeek V4.1 Flash measured |
| LMArena Vision: reading text | 1282 | Not measured | Only DeepSeek V4.1 Flash measured |
| LMArena Vision: humor | 1260 | Not measured | Only DeepSeek V4.1 Flash measured |
| LMArena Vision: creative writing | 1275 | Not measured | Only DeepSeek V4.1 Flash measured |
| 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 | Gemini 4 Argon | 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 | Google Gemini 4 Argon |
|---|---|---|
| Maker | DeepSeek | |
| Weights | Open weights | No public weights listed |
| Released | September 10, 2026 | October 13, 2026 |
| Context window | 1.05M | 1.05M |
| Input $/Mtok | $0.2225 | $2.00 |
| Output $/Mtok | $0.84 | $10.00 |
| Coding index | 62.5 | 81.0 |
| Independent measurements behind it | 5 | 5 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is DeepSeek V4.1 Flash better than Gemini 4 Argon for coding?
Pick Gemini 4 Argon for coding: it scores 81.0 on the Vibe Coding Index against 62.5 for DeepSeek V4.1 Flash. Pick DeepSeek V4.1 Flash for output tokens 11.9x cheaper ($0.84 vs $10.00 per million) and work measured by LiveBench coding or Terminal-Bench 2.1 (Vals AI), where it leads.
Which is cheaper, DeepSeek V4.1 Flash or Gemini 4 Argon?
DeepSeek V4.1 Flash costs $0.2225 per million input tokens and $0.84 per million output tokens; Gemini 4 Argon costs $2.00 and $10.00. Cost per solved task is known for neither of them.
Which has the bigger context window, DeepSeek V4.1 Flash or Gemini 4 Argon?
Neither: both take 1.05M tokens.
How many benchmarks is the DeepSeek V4.1 Flash vs Gemini 4 Argon comparison based on?
83 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 4 Argon's on 5. Where only one model has a result, the page says so instead of naming a winner.