DeepSeek V4 Flash 0731 vs DeepSeek V4 Pro 0423
DeepSeek V4 Flash 0731 leads DeepSeek V4 Pro 0423 on coding, 51.8 to 38.7 on the Vibe Coding Index, and DeepSeek V4 Flash 0731 costs 9.7x less per output token ($0.28 vs $2.716 per million). Updated October 21, 2026; 44 benchmarks have a result for both.
DeepSeek V4 Flash 0731 or DeepSeek V4 Pro 0423: which to pick
Pick DeepSeek V4 Flash 0731 for coding: it scores 51.8 on the Vibe Coding Index against 38.7 for DeepSeek V4 Pro 0423, though on 4 benchmarks against 5.
Pick DeepSeek V4 Pro 0423 for work measured by LiveCodeBench (Vals AI) or SkillsBench (Vals AI), where it leads.
- Shared results
- 44
- DeepSeek V4 Flash 0731 only
- 11
- DeepSeek V4 Pro 0423 only
- 40
- 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 Flash 0731
DeepSeek V4 Flash 0731 scores 51.8 against 38.7.
4 and 5 benchmarks behind the scores
Intelligence
AA Intelligence Index
DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 leads, 34.3 against 30.4.
Same board, both measured
Agentic work
AA Agentic Index
DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 leads, 41.0 against 26.3.
Same board, both measured
Reasoning
GPQA Diamond
DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 leads, 89.9% against 89.4%.
Same board, both measured
Knowledge
MMLU Pro
DeepSeek V4 Pro 0423
DeepSeek V4 Pro 0423 leads, 87.3% against 86.2%.
Same board, both measured
Web design
Design Arena: websites
DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 leads, 1238 against 1234.
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 Flash 0731
DeepSeek V4 Flash 0731 $0.0003, DeepSeek V4 Pro 0423 $0.0026 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 $0.01, DeepSeek V4 Pro 0423 $0.57.
Published attempt cost and pass rate
At a glance
- Coding index
- DeepSeek V4 Flash 0731 51.8 from 4 benchmarks, DeepSeek V4 Pro 0423 38.7 from 5 benchmarks.
- Widest category
- Math, 19.1 points apart on average over 4 shared benchmarks.
- Widest single gap
- ProofBench: DeepSeek V4 Flash 0731 56.0% against DeepSeek V4 Pro 0423 16.0%, 40.0 points apart on a 0 to 100 scale.
- Output price
- DeepSeek V4 Flash 0731 $0.28 against DeepSeek V4 Pro 0423 $2.716 per million output tokens, 9.7x 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 Flash 0731 | Bars | DeepSeek V4 Pro 0423 | Ties |
|---|---|---|---|---|
| Coding | 9 | 2 | 1 tie | |
| Intelligence | 2 | 0 | ||
| Agentic work | 5 | 0 | ||
| Reasoning | 4 | 0 | ||
| Math | 3 | 1 | ||
| Knowledge | 2 | 4 | ||
| Web design | 3 | 5 | ||
| General | 3 | 0 | ||
| All shared boards | 31 | 12 | 1 tie |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index92
| Benchmark | DeepSeek V4 Flash 0731 | DeepSeek V4 Pro 0423 | Leader |
|---|---|---|---|
| LMArena WebDev | Not measured | 1447 | Only DeepSeek V4 Pro 0423 measured |
| AA Coding Index | 69.1 | 59.4 | DeepSeek V4 Flash 0731 |
| LiveCodeBench (Vals AI) | 87.3% | 87.5% | DeepSeek V4 Pro 0423 |
| SWE-bench Verified (Vals AI) | 88.8% | 77.4% | DeepSeek V4 Flash 0731 |
| Terminal-Bench 4 (Vals AI) | 18.7% | 11.1% | DeepSeek V4 Flash 0731 |
| Vibe Code Bench (Vals AI) | 74.7% | 49.9% | DeepSeek V4 Flash 0731 |
| Code migration (Vals AI) | 38.6% | 26.2% | DeepSeek V4 Flash 0731 |
| ProgramBench (Vals AI) | 0.0% | 0.0% | Tie |
| IOI (Vals AI) | 32.7% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Terminal-Bench 2.1 (Vals AI) | 67.0% | 50.2% | DeepSeek V4 Flash 0731 |
| SkillsBench (Vals AI) | 50.7% | 51.3% | DeepSeek V4 Pro 0423 |
| CyberBench (Vals AI) | 72.1% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| SciCode | 50.3% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| FrontierCode | 18.8% | 17.6% | DeepSeek V4 Flash 0731 |
| WeirdML | 63.0% | 48.9% | DeepSeek V4 Flash 0731 |
| ALE-Bench | 1306.1 | 1006.1 | DeepSeek V4 Flash 0731 |
| LMArena Text: coding | Not measured | 1470 | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: software and IT | Not measured | 1469 | Only DeepSeek V4 Pro 0423 measured |
| LMArena WebDev: React | Not measured | 1441 | Only DeepSeek V4 Pro 0423 measured |
| LMArena WebDev: HTML | Not measured | 1435 | Only DeepSeek V4 Pro 0423 measured |
| Neither model has a result on the other 19 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | DeepSeek V4 Flash 0731 | DeepSeek V4 Pro 0423 | Leader |
|---|---|---|---|
| AA Intelligence Index | 34.3 | 30.4 | DeepSeek V4 Flash 0731 |
| Epoch Capabilities Index | 154.8 | 149.1 | DeepSeek V4 Flash 0731 |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index50
| Benchmark | DeepSeek V4 Flash 0731 | DeepSeek V4 Pro 0423 | Leader |
|---|---|---|---|
| AA Agentic Index | 41.0 | 26.3 | DeepSeek V4 Flash 0731 |
| Design Arena: agents building web apps | Not measured | 1000 | Only DeepSeek V4 Pro 0423 measured |
| Design Arena: agents building full-stack apps | Not measured | 948 | Only DeepSeek V4 Pro 0423 measured |
| Design Arena: agents building Godot games | Not measured | 1059 | Only DeepSeek V4 Pro 0423 measured |
| Finance Agent (Vals AI) | 49.5% | 44.1% | DeepSeek V4 Flash 0731 |
| Tax Agent Bench (Vals AI) | 28.1% | 25.2% | DeepSeek V4 Flash 0731 |
| Legal Research Bench (Vals AI) | 30.3% | 23.1% | DeepSeek V4 Flash 0731 |
| Excel Modeling Benchmark (Vals AI) | 57.0% | 51.6% | DeepSeek V4 Flash 0731 |
| 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 |
| Vending-Bench 2 | Not measured | $3,285 | Only DeepSeek V4 Pro 0423 measured |
| Neither model has a result on the other 30 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index40
| Benchmark | DeepSeek V4 Flash 0731 | DeepSeek V4 Pro 0423 | Leader |
|---|---|---|---|
| GPQA Diamond | 89.9% | 89.4% | DeepSeek V4 Flash 0731 |
| GPQA Diamond (Artificial Analysis run) | 90.8% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Humanity's Last Exam (Artificial Analysis run) | 38.6% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| GPQA Diamond (Epoch AI run) | 91.0% | 90.9% | DeepSeek V4 Flash 0731 |
| ARC-AGI | 89.0% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| ARC-AGI-2 | 61.4% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| SimpleBench | 61.1% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Chess puzzles | 32.0% | 19.0% | DeepSeek V4 Flash 0731 |
| Mystery game puzzles | 34.0% | 17.0% | DeepSeek V4 Flash 0731 |
| LMArena Text: hard prompts | Not measured | 1461 | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: expert | Not measured | 1463 | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: hard prompts in English | Not measured | 1466 | Only DeepSeek V4 Pro 0423 measured |
| Kagi LLM Benchmark | Not measured | 53.5% | Only DeepSeek V4 Pro 0423 measured |
| Neither model has a result on the other 9 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index31
| Benchmark | DeepSeek V4 Flash 0731 | DeepSeek V4 Pro 0423 | Leader |
|---|---|---|---|
| OTIS Mock AIME | 94.4% | 96.7% | DeepSeek V4 Pro 0423 |
| FrontierMath, tiers 1 to 3 | 57.5% | 45.3% | DeepSeek V4 Flash 0731 |
| FrontierMath, tier 4 | 24.4% | 2.4% | DeepSeek V4 Flash 0731 |
| ProofBench | 56.0% | 16.0% | DeepSeek V4 Flash 0731 |
| LMArena Text: math | Not measured | 1443 | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: mathematical work | Not measured | 1455 | Only DeepSeek V4 Pro 0423 measured |
| Neither model has a result on the other 7 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index24
| Benchmark | DeepSeek V4 Flash 0731 | DeepSeek V4 Pro 0423 | Leader |
|---|---|---|---|
| MMLU Pro | 86.2% | 87.3% | DeepSeek V4 Pro 0423 |
| LegalBench (Vals AI) | 77.7% | 80.3% | DeepSeek V4 Pro 0423 |
| CaseLaw (Vals AI) | Not measured | 59.4% | Only DeepSeek V4 Pro 0423 measured |
| CorpFin (Vals AI) | 61.9% | 61.4% | DeepSeek V4 Flash 0731 |
| TaxEval (Vals AI) | 70.7% | 72.1% | DeepSeek V4 Pro 0423 |
| MedCode (Vals AI) | 41.4% | 40.5% | DeepSeek V4 Flash 0731 |
| SimpleQA Verified | 33.6% | 47.0% | DeepSeek V4 Pro 0423 |
| LMArena Text: science | Not measured | 1470 | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: medicine and healthcare | Not measured | 1466 | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: legal and government | Not measured | 1464 | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: business and finance | Not measured | 1443 | Only DeepSeek V4 Pro 0423 measured |
| Neither model has a result on the other 7 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index35
| Benchmark | DeepSeek V4 Flash 0731 | DeepSeek V4 Pro 0423 | Leader |
|---|---|---|---|
| Design Arena: websites | 1238 | 1234 | DeepSeek V4 Flash 0731 |
| Design Arena: UI components | 1235 | 1222 | DeepSeek V4 Flash 0731 |
| Design Arena: game dev | 1208 | 1229 | DeepSeek V4 Pro 0423 |
| Design Arena: data viz | 1180 | 1203 | DeepSeek V4 Pro 0423 |
| Design Arena: 3D | 1209 | 1254 | DeepSeek V4 Pro 0423 |
| Design Arena: all code categories | 1230 | 1240 | DeepSeek V4 Pro 0423 |
| Design Arena: SVG | 1185 | 1140 | DeepSeek V4 Flash 0731 |
| Design Arena: ASCII art | 1090 | 1149 | DeepSeek V4 Pro 0423 |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index30
| Benchmark | DeepSeek V4 Flash 0731 | DeepSeek V4 Pro 0423 | Leader |
|---|---|---|---|
| Public Benefits Bench (Vals AI) | Not measured | 62.9% | Only DeepSeek V4 Pro 0423 measured |
| MedScribe (Vals AI) | 80.4% | 75.1% | DeepSeek V4 Flash 0731 |
| Long context reasoning (AA-LCR) | 79.7% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| DTBench | 90.9% | 90.7% | DeepSeek V4 Flash 0731 |
| LMCA | 41.7% | 41.2% | DeepSeek V4 Flash 0731 |
| Surface Evolver Bench | Not measured | 40.0% | Only DeepSeek V4 Pro 0423 measured |
| CL-bench Life | Not measured | 13.5% | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text | Not measured | 1451 | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: instruction following | Not measured | 1445 | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: creative writing | Not measured | 1446 | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: English | Not measured | 1459 | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: non-English | Not measured | 1439 | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: Chinese | Not measured | 1481 | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: German | Not measured | 1458 | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: Japanese | Not measured | 1446 | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: Korean | Not measured | 1409 | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: French | Not measured | 1473 | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: Spanish | Not measured | 1457 | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: Russian | Not measured | 1453 | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: Polish | Not measured | 1461 | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: multi-turn | Not measured | 1468 | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: longer queries | Not measured | 1458 | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: writing and language | Not measured | 1447 | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: entertainment and media | Not measured | 1431 | Only DeepSeek V4 Pro 0423 measured |
| Neither model has a result on the other 22 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | DeepSeek V4 Flash 0731 | DeepSeek V4 Pro 0423 | 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 | DeepSeek V4 Pro 0423 |
|---|---|---|
| Maker | DeepSeek | DeepSeek |
| Weights | Open weights | Open weights |
| Released | July 31, 2026 | April 24, 2026 |
| Context window | 1.05M | 1.05M |
| Input $/Mtok | $0.14 | $1.225 |
| Output $/Mtok | $0.28 | $2.716 |
| Coding index | 51.8 | 38.7 |
| Independent measurements behind it | 4 | 5 |
| Cost per solved task | $0.01 | $0.57 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is DeepSeek V4 Flash 0731 better than DeepSeek V4 Pro 0423 for coding?
Pick DeepSeek V4 Flash 0731 for coding: it scores 51.8 on the Vibe Coding Index against 38.7 for DeepSeek V4 Pro 0423, though on 4 benchmarks against 5. Pick DeepSeek V4 Pro 0423 for work measured by LiveCodeBench (Vals AI) or SkillsBench (Vals AI), where it leads.
Which is cheaper, DeepSeek V4 Flash 0731 or DeepSeek V4 Pro 0423?
DeepSeek V4 Flash 0731 costs $0.14 per million input tokens and $0.28 per million output tokens; DeepSeek V4 Pro 0423 costs $1.225 and $2.716. Per solved SWE-bench Verified task, DeepSeek V4 Flash 0731 costs $0.01 and DeepSeek V4 Pro 0423 $0.57.
Which has the bigger context window, DeepSeek V4 Flash 0731 or DeepSeek V4 Pro 0423?
Neither: both take 1.05M tokens.
How many benchmarks is the DeepSeek V4 Flash 0731 vs DeepSeek V4 Pro 0423 comparison based on?
44 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 DeepSeek V4 Pro 0423's on 5. Where only one model has a result, the page says so instead of naming a winner.