DeepSeek V4 Flash 0731 vs o3
DeepSeek V4 Flash 0731 leads o3 on the Artificial Analysis Intelligence Index, 34.3 to 20.2, and DeepSeek V4 Flash 0731 costs 28.6x less per output token ($0.28 vs $8.00 per million). Updated October 21, 2026; 31 benchmarks have a result for both.
DeepSeek V4 Flash 0731 or o3: which to pick
Only DeepSeek V4 Flash 0731 has a coding score, 51.8 on the Vibe Coding Index; o3 has none, so the two cannot be ranked on coding.
Pick o3 for work measured by Chess puzzles or LegalBench (Vals AI), where it leads.
- Shared results
- 31
- DeepSeek V4 Flash 0731 only
- 24
- o3 only
- 66
- Like-for-like categories
- 7 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
Only DeepSeek V4 Flash 0731 has a coding score, so no call is made.
One side not measured
Intelligence
AA Intelligence Index
DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 leads, 34.3 against 20.2.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only DeepSeek V4 Flash 0731 has a AA Agentic Index result, 41.0.
Directional only
Reasoning
GPQA Diamond
DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 leads, 89.9% against 84.1%.
Same board, both measured
Math
AIME
No call
Only o3 has a AIME result, 85.3%.
Directional only
Knowledge
MMLU Pro
DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 leads, 86.2% against 85.6%.
Same board, both measured
Web design
Design Arena: websites
DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 leads, 1238 against 1036.
Same board, both measured
Long documents
Listed context window
DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 takes 1.05M tokens against 200K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 $0.0003, o3 $0.0060 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 $0.01, o3 $0.571.
Published attempt cost and pass rate
At a glance
- Coding index
- DeepSeek V4 Flash 0731 51.8 from 4 benchmarks, o3 not benchmarked.
- Widest category
- Web design, 20.6 points apart on average over 5 shared benchmarks.
- Widest single gap
- ARC-AGI-2: DeepSeek V4 Flash 0731 61.4% against o3 6.5%, 54.9 points apart on a 0 to 100 scale.
- Output price
- DeepSeek V4 Flash 0731 $0.28 against o3 $8.00 per million output tokens, 28.6x apart.
- Context window
- DeepSeek V4 Flash 0731 takes 1.05M tokens, o3 200K.
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 Flash 0731 | Bars | o3 | Ties |
|---|---|---|---|---|
| Coding | 3 | 0 | ||
| Intelligence | 2 | 0 | ||
| Reasoning | 8 | 1 | ||
| Math | 2 | 0 | ||
| Knowledge | 2 | 4 | ||
| Web design | 4 | 1 | ||
| General | 4 | 0 | ||
| All shared boards | 25 | 6 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index30
| Benchmark | DeepSeek V4 Flash 0731 | o3 | Leader |
|---|---|---|---|
| Aider polyglot | Not measured | 81.3% | Only o3 measured |
| SWE-bench Verified | Not measured | 58.4% | Only o3 measured |
| SWE-bench Multimodal | Not measured | 36.0% | Only o3 measured |
| AA Coding Index | 69.1 | Not measured | Only DeepSeek V4 Flash 0731 measured |
| LiveCodeBench (Vals AI) | 87.3% | 83.9% | DeepSeek V4 Flash 0731 |
| SWE-bench Verified (Vals AI) | 88.8% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Terminal-Bench 4 (Vals AI) | 18.7% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Vibe Code Bench (Vals AI) | 74.7% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Code migration (Vals AI) | 38.6% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| ProgramBench (Vals AI) | 0.0% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| IOI (Vals AI) | 32.7% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Terminal-Bench 2.1 (Vals AI) | 67.0% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| SkillsBench (Vals AI) | 50.7% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| CyberBench (Vals AI) | 72.1% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| LiveCodeBench (Artificial Analysis run) | Not measured | 80.8% | Only o3 measured |
| SciCode | 50.3% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 37.1% | Only o3 measured |
| FrontierCode | 18.8% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| GSO-Bench | Not measured | 8.8% | Only o3 measured |
| WeirdML | 63.0% | 52.4% | DeepSeek V4 Flash 0731 |
| ALE-Bench | 1306.1 | 933.5 | DeepSeek V4 Flash 0731 |
| LMArena Text: coding | Not measured | 1407 | Only o3 measured |
| LMArena Text: software and IT | Not measured | 1415 | Only o3 measured |
| Neither model has a result on the other 15 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | DeepSeek V4 Flash 0731 | o3 | Leader |
|---|---|---|---|
| AA Intelligence Index | 34.3 | 20.2 | DeepSeek V4 Flash 0731 |
| Epoch Capabilities Index | 154.8 | 146.7 | DeepSeek V4 Flash 0731 |
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index12 boards, none shared
| Benchmark | DeepSeek V4 Flash 0731 | o3 | Leader |
|---|---|---|---|
| AA Agentic Index | 41.0 | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Finance Agent (Vals AI) | 49.5% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Tax Agent Bench (Vals AI) | 28.1% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Legal Research Bench (Vals AI) | 30.3% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Excel Modeling Benchmark (Vals AI) | 57.0% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| BioMysteryBench (Vals AI) | 64.4% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Terminal-Bench-Science (Vals AI) | 4.3% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| τ²-Bench Telecom | Not measured | 80.7% | Only o3 measured |
| METR time horizons | Not measured | 65.4% | Only o3 measured |
| GDPval | Not measured | 30.8% | Only o3 measured |
| DeepResearch Bench | Not measured | 45.2% | Only o3 measured |
| OSWorld | Not measured | 23.0% | Only o3 measured |
| Neither model has a result on the other 27 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index81
| Benchmark | DeepSeek V4 Flash 0731 | o3 | Leader |
|---|---|---|---|
| GPQA Diamond | 89.9% | 84.1% | DeepSeek V4 Flash 0731 |
| MMMU Pro (Vals AI) | Not measured | 80.4% | Only o3 measured |
| GPQA Diamond (Artificial Analysis run) | 90.8% | 82.7% | DeepSeek V4 Flash 0731 |
| Humanity's Last Exam (Artificial Analysis run) | 38.6% | 20.1% | DeepSeek V4 Flash 0731 |
| GPQA Diamond (Epoch AI run) | 91.0% | 81.8% | DeepSeek V4 Flash 0731 |
| ARC-AGI | 89.0% | 60.8% | DeepSeek V4 Flash 0731 |
| ARC-AGI-2 | 61.4% | 6.5% | DeepSeek V4 Flash 0731 |
| SimpleBench | 61.1% | 53.1% | DeepSeek V4 Flash 0731 |
| Chess puzzles | 32.0% | 37.0% | o3 |
| Mystery game puzzles | 34.0% | 29.0% | DeepSeek V4 Flash 0731 |
| VPCT | Not measured | 52.0% | Only o3 measured |
| LMArena Text: hard prompts | Not measured | 1402 | Only o3 measured |
| LMArena Text: expert | Not measured | 1402 | Only o3 measured |
| LMArena Text: hard prompts in English | Not measured | 1411 | Only o3 measured |
| LMArena Vision: diagrams | Not measured | 1220 | Only o3 measured |
| LMArena Vision: homework | Not measured | 1234 | Only o3 measured |
| Neither model has a result on the other 5 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | DeepSeek V4 Flash 0731 | o3 | Leader |
|---|---|---|---|
| AIME | Not measured | 85.3% | Only o3 measured |
| MATH 500 (Vals AI) | Not measured | 94.6% | Only o3 measured |
| MGSM (Vals AI) | Not measured | 91.8% | Only o3 measured |
| AA Math Index | Not measured | 88.3 | Only o3 measured |
| AIME 2025 (Artificial Analysis run) | Not measured | 88.3% | Only o3 measured |
| OTIS Mock AIME | 94.4% | 84.4% | DeepSeek V4 Flash 0731 |
| FrontierMath, tiers 1 to 3 | 57.5% | 33.3% | DeepSeek V4 Flash 0731 |
| FrontierMath, tier 4 | 24.4% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| ProofBench | 56.0% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| MATH level 5 | Not measured | 97.8% | Only o3 measured |
| LMArena Text: math | Not measured | 1426 | Only o3 measured |
| LMArena Text: mathematical work | Not measured | 1423 | Only o3 measured |
| Neither model has a result on the other 1 board here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index24
| Benchmark | DeepSeek V4 Flash 0731 | o3 | Leader |
|---|---|---|---|
| MMLU Pro | 86.2% | 85.6% | DeepSeek V4 Flash 0731 |
| LegalBench (Vals AI) | 77.7% | 83.8% | o3 |
| CorpFin (Vals AI) | 61.9% | 59.7% | DeepSeek V4 Flash 0731 |
| TaxEval (Vals AI) | 70.7% | 74.6% | o3 |
| MortgageTax (Vals AI) | Not measured | 65.7% | Only o3 measured |
| MedQA (Vals AI) | Not measured | 96.1% | Only o3 measured |
| MedCode (Vals AI) | 41.4% | 47.3% | o3 |
| MMLU Pro (Artificial Analysis run) | Not measured | 85.3% | Only o3 measured |
| SimpleQA Verified | 33.6% | 49.4% | o3 |
| GeoBench | Not measured | 74.0% | Only o3 measured |
| LMArena Text: science | Not measured | 1429 | Only o3 measured |
| LMArena Text: medicine and healthcare | Not measured | 1448 | Only o3 measured |
| LMArena Text: legal and government | Not measured | 1428 | Only o3 measured |
| LMArena Text: business and finance | Not measured | 1393 | Only o3 measured |
| Neither model has a result on the other 3 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index41
| Benchmark | DeepSeek V4 Flash 0731 | o3 | Leader |
|---|---|---|---|
| Design Arena: websites | 1238 | 1036 | DeepSeek V4 Flash 0731 |
| Design Arena: UI components | 1235 | 1017 | DeepSeek V4 Flash 0731 |
| Design Arena: game dev | 1208 | 1034 | DeepSeek V4 Flash 0731 |
| Design Arena: data viz | 1180 | 1200 | o3 |
| Design Arena: 3D | 1209 | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Design Arena: all code categories | 1230 | 1020 | DeepSeek V4 Flash 0731 |
| Design Arena: SVG | 1185 | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Design Arena: ASCII art | 1090 | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index40
| Benchmark | DeepSeek V4 Flash 0731 | o3 | Leader |
|---|---|---|---|
| SAGE (Vals AI) | Not measured | 41.8% | Only o3 measured |
| MedScribe (Vals AI) | 80.4% | 76.7% | DeepSeek V4 Flash 0731 |
| IFBench | Not measured | 71.4% | Only o3 measured |
| Long context reasoning (AA-LCR) | 79.7% | 74.7% | DeepSeek V4 Flash 0731 |
| DTBench | 90.9% | 84.8% | DeepSeek V4 Flash 0731 |
| LMCA | 41.7% | 39.7% | DeepSeek V4 Flash 0731 |
| Fiction.LiveBench | Not measured | 88.9% | Only o3 measured |
| Creative writing (Lech Mazur) | Not measured | 83.9% | Only o3 measured |
| CL-bench | Not measured | 17.8% | Only o3 measured |
| LMArena Text | Not measured | 1410 | Only o3 measured |
| LMArena Text: instruction following | Not measured | 1368 | Only o3 measured |
| LMArena Text: creative writing | Not measured | 1359 | Only o3 measured |
| LMArena Vision | Not measured | 1214 | Only o3 measured |
| LMArena Text: English | Not measured | 1417 | Only o3 measured |
| LMArena Text: non-English | Not measured | 1401 | Only o3 measured |
| LMArena Text: Chinese | Not measured | 1437 | Only o3 measured |
| LMArena Text: German | Not measured | 1420 | Only o3 measured |
| LMArena Text: Japanese | Not measured | 1403 | Only o3 measured |
| LMArena Text: Korean | Not measured | 1370 | Only o3 measured |
| LMArena Text: French | Not measured | 1430 | Only o3 measured |
| LMArena Text: Spanish | Not measured | 1395 | Only o3 measured |
| LMArena Text: Russian | Not measured | 1406 | Only o3 measured |
| LMArena Text: Polish | Not measured | 1438 | Only o3 measured |
| LMArena Text: multi-turn | Not measured | 1405 | Only o3 measured |
| LMArena Text: longer queries | Not measured | 1372 | Only o3 measured |
| LMArena Text: writing and language | Not measured | 1372 | Only o3 measured |
| LMArena Text: entertainment and media | Not measured | 1373 | Only o3 measured |
| LMArena Vision: English | Not measured | 1221 | Only o3 measured |
| LMArena Vision: Chinese | Not measured | 1237 | Only o3 measured |
| LMArena Vision: reading text | Not measured | 1218 | Only o3 measured |
| LMArena Vision: humor | Not measured | 1222 | Only o3 measured |
| LMArena Vision: captioning | Not measured | 1210 | Only o3 measured |
| LMArena Vision: recognizing things | Not measured | 1234 | Only o3 measured |
| LMArena Vision: creative writing | Not measured | 1201 | Only o3 measured |
| Neither model has a result on the other 12 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | DeepSeek V4 Flash 0731 | o3 | 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 Flash 0731 costs less
Repository review
50K input + 3K output tokens
DeepSeek V4 Flash 0731 costs less
Agent loop
200K input + 10K output tokens
DeepSeek V4 Flash 0731 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 Flash 0731 | OpenAI o3 |
|---|---|---|
| Maker | DeepSeek | OpenAI |
| Weights | Open weights | No public weights listed |
| Released | July 31, 2026 | April 16, 2025 |
| Context window | 1.05M | 200K |
| Input $/Mtok | $0.14 | $2.00 |
| Output $/Mtok | $0.28 | $8.00 |
| Coding index | 51.8 | Not benchmarked |
| Independent measurements behind it | 4 | 0 |
| Cost per solved task | $0.01 | $0.571 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is DeepSeek V4 Flash 0731 better than o3 for coding?
Only DeepSeek V4 Flash 0731 has a coding score, 51.8 on the Vibe Coding Index; o3 has none, so the two cannot be ranked on coding. Pick o3 for work measured by Chess puzzles or LegalBench (Vals AI), where it leads.
Which is cheaper, DeepSeek V4 Flash 0731 or o3?
DeepSeek V4 Flash 0731 costs $0.14 per million input tokens and $0.28 per million output tokens; o3 costs $2.00 and $8.00. Per solved SWE-bench Verified task, DeepSeek V4 Flash 0731 costs $0.01 and o3 $0.571.
Which has the bigger context window, DeepSeek V4 Flash 0731 or o3?
DeepSeek V4 Flash 0731, with 1.05M tokens against 200K.
How many benchmarks is the DeepSeek V4 Flash 0731 vs o3 comparison based on?
31 benchmarks have a result for both models as of October 21, 2026. DeepSeek V4 Flash 0731's coding index rests on 4 independent benchmarks and o3's on 0. Where only one model has a result, the page says so instead of naming a winner.