DeepSeek V4 Pro 0813 vs GPT-5.2
DeepSeek V4 Pro 0813 leads GPT-5.2 on coding, 56.1 to 55.9 on the Vibe Coding Index, and DeepSeek V4 Pro 0813 costs 7.1x less per output token ($1.98 vs $14.00 per million). Updated October 22, 2026; 56 benchmarks have a result for both.
DeepSeek V4 Pro 0813 or GPT-5.2: which to pick
DeepSeek V4 Pro 0813 and GPT-5.2 are level for coding, 56.1 against 55.9 on the Vibe Coding Index, so price and context decide.
Pick GPT-5.2 for work measured by WeirdML or Chess puzzles, where it leads. Pick DeepSeek V4 Pro 0813 for output tokens 7.1x cheaper ($1.98 vs $14.00 per million) and a lower cost per solved task ($0.10 against $1.91).
- Shared results
- 56
- DeepSeek V4 Pro 0813 only
- 22
- GPT-5.2 only
- 58
- Like-for-like categories
- 6 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: 56.1 against 55.9, inside 2 points.
5 and 3 benchmarks behind the scores
Intelligence
AA Intelligence Index
DeepSeek V4 Pro 0813
DeepSeek V4 Pro 0813 leads, 36.0 against 30.4.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only DeepSeek V4 Pro 0813 has a AA Agentic Index result, 41.3.
Directional only
Reasoning
GPQA Diamond
DeepSeek V4 Pro 0813
DeepSeek V4 Pro 0813 leads, 92.4% against 91.7%.
Same board, both measured
Math
AIME
No call
Only GPT-5.2 has a AIME result, 96.9%.
Directional only
Knowledge
MMLU Pro
DeepSeek V4 Pro 0813
DeepSeek V4 Pro 0813 leads, 87.0% against 86.2%.
Same board, both measured
Web design
Design Arena: websites
No call
Only GPT-5.2 has a Design Arena: websites result, 1195.
Directional only
General
Vals Index
No call
Only DeepSeek V4 Pro 0813 has a Vals Index result, 53.0%.
Directional only
Long documents
Listed context window
DeepSeek V4 Pro 0813
DeepSeek V4 Pro 0813 takes 1.05M tokens against 400K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
DeepSeek V4 Pro 0813
DeepSeek V4 Pro 0813 $0.0016, GPT-5.2 $0.0088 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
DeepSeek V4 Pro 0813
DeepSeek V4 Pro 0813 $0.10, GPT-5.2 $1.91.
Published attempt cost and pass rate
At a glance
- Coding index
- DeepSeek V4 Pro 0813 56.1 from 5 benchmarks, GPT-5.2 55.9 from 3 benchmarks.
- Widest category
- Coding, 13.0 points apart on average over 9 shared benchmarks.
- Widest single gap
- ProofBench: DeepSeek V4 Pro 0813 50.0% against GPT-5.2 15.0%, 35.0 points apart on a 0 to 100 scale.
- Output price
- DeepSeek V4 Pro 0813 $1.98 against GPT-5.2 $14.00 per million output tokens, 7.1x apart.
- Context window
- DeepSeek V4 Pro 0813 takes 1.05M tokens, GPT-5.2 400K.
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 Pro 0813 | Bars | GPT-5.2 | Ties |
|---|---|---|---|---|
| Coding | 8 | 1 | ||
| Intelligence | 2 | 0 | ||
| Reasoning | 10 | 1 | ||
| Math | 4 | 2 | ||
| Knowledge | 6 | 4 | ||
| General | 15 | 3 | ||
| All shared boards | 45 | 11 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index81
| Benchmark | DeepSeek V4 Pro 0813 | GPT-5.2 | Leader |
|---|---|---|---|
| SWE-bench Verified | Not measured | 72.8% | Only GPT-5.2 measured |
| SWE-bench Multilingual | Not measured | 66.7% | Only GPT-5.2 measured |
| LMArena WebDev | 1582 | 1417 | DeepSeek V4 Pro 0813 |
| AA Coding Index | 68.8 | Not measured | Only DeepSeek V4 Pro 0813 measured |
| LiveCodeBench (Vals AI) | 87.5% | 85.4% | DeepSeek V4 Pro 0813 |
| SWE-bench Verified (Vals AI) | 96.4% | 75.8% | DeepSeek V4 Pro 0813 |
| Terminal-Bench 4 (Vals AI) | 14.1% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| Vibe Code Bench (Vals AI) | 82.3% | 53.5% | DeepSeek V4 Pro 0813 |
| Code migration (Vals AI) | 41.5% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| ProgramBench (Vals AI) | 0.0% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| IOI (Vals AI) | 51.6% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| Terminal-Bench 2.1 (Vals AI) | 54.7% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| Vibe Code Bench 1-100 (Vals AI) | 17.5% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| SkillsBench (Vals AI) | 53.8% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| CyberBench (Vals AI) | 72.9% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| LiveCodeBench (Artificial Analysis run) | Not measured | 89.4% | Only GPT-5.2 measured |
| SciCode | 51.0% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 47.0% | Only GPT-5.2 measured |
| Terminal-Bench | Not measured | 64.9% | Only GPT-5.2 measured |
| FrontierCode | 28.5% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| GSO-Bench | Not measured | 27.5% | Only GPT-5.2 measured |
| WeirdML | 66.2% | 72.2% | GPT-5.2 |
| ALE-Bench | 1403.2 | 1293.5 | DeepSeek V4 Pro 0813 |
| LMArena Text: coding | 1469 | 1442 | DeepSeek V4 Pro 0813 |
| LMArena Text: software and IT | 1462 | 1434 | DeepSeek V4 Pro 0813 |
| LMArena WebDev: React | 1588 | Not measured | Only DeepSeek V4 Pro 0813 measured |
| LMArena WebDev: HTML | 1566 | 1427 | DeepSeek V4 Pro 0813 |
| Neither model has a result on the other 12 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | DeepSeek V4 Pro 0813 | GPT-5.2 | Leader |
|---|---|---|---|
| AA Intelligence Index | 36.0 | 30.4 | DeepSeek V4 Pro 0813 |
| Epoch Capabilities Index | 155.6 | 153.7 | DeepSeek V4 Pro 0813 |
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index21 boards, none shared
| Benchmark | DeepSeek V4 Pro 0813 | GPT-5.2 | Leader |
|---|---|---|---|
| AA Agentic Index | 41.3 | Not measured | Only DeepSeek V4 Pro 0813 measured |
| Design Arena: agents building web apps | Not measured | 1086 | Only GPT-5.2 measured |
| Design Arena: agents building full-stack apps | Not measured | 1043 | Only GPT-5.2 measured |
| Design Arena: agents building mobile apps | Not measured | 1093 | Only GPT-5.2 measured |
| Design Arena: agents building Android apps | Not measured | 1059 | Only GPT-5.2 measured |
| Design Arena: agents building Godot games | Not measured | 1142 | Only GPT-5.2 measured |
| SREBench (Vals AI) | 1.9% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| Finance Agent (Vals AI) | 50.4% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| Tax Agent Bench (Vals AI) | 26.8% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| Legal Research Bench (Vals AI) | 40.9% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| Harvey's Legal Agent Benchmark (Vals AI) | 10.8% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| Excel Modeling Benchmark (Vals AI) | 52.8% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| Terminal-Bench-Science (Vals AI) | 4.3% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| τ²-Bench Telecom | Not measured | 84.8% | Only GPT-5.2 measured |
| APEX-Agents | 47.3% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| METR time horizons | Not measured | 75.3% | Only GPT-5.2 measured |
| Remote Labor Index | Not measured | 2.5% | Only GPT-5.2 measured |
| GDPval | Not measured | 49.7% | Only GPT-5.2 measured |
| DeepResearch Bench | Not measured | 41.1% | Only GPT-5.2 measured |
| Vending-Bench 2 | Not measured | $3,591 | Only GPT-5.2 measured |
| Berkeley Function Calling Leaderboard (BFCL) | Not measured | 55.9% | Only GPT-5.2 measured |
| Neither model has a result on the other 20 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index101
| Benchmark | DeepSeek V4 Pro 0813 | GPT-5.2 | Leader |
|---|---|---|---|
| GPQA Diamond | 92.4% | 91.7% | DeepSeek V4 Pro 0813 |
| MMMU Pro (Vals AI) | Not measured | 86.7% | Only GPT-5.2 measured |
| GPQA Diamond (Artificial Analysis run) | 92.8% | 90.3% | DeepSeek V4 Pro 0813 |
| Humanity's Last Exam (Artificial Analysis run) | 41.0% | 37.7% | DeepSeek V4 Pro 0813 |
| GPQA Diamond (Epoch AI run) | 91.7% | 91.4% | DeepSeek V4 Pro 0813 |
| ARC-AGI | 90.5% | 86.2% | DeepSeek V4 Pro 0813 |
| ARC-AGI-2 | 61.3% | 52.9% | DeepSeek V4 Pro 0813 |
| SimpleBench | Not measured | 45.8% | Only GPT-5.2 measured |
| Chess puzzles | 46.0% | 48.0% | GPT-5.2 |
| Mystery game puzzles | 43.0% | 23.0% | DeepSeek V4 Pro 0813 |
| EBR-bench | Not measured | 23.0% | Only GPT-5.2 measured |
| VPCT | Not measured | 84.0% | Only GPT-5.2 measured |
| LMArena Text: hard prompts | 1458 | 1428 | DeepSeek V4 Pro 0813 |
| LMArena Text: expert | 1459 | 1445 | DeepSeek V4 Pro 0813 |
| LMArena Text: hard prompts in English | 1463 | 1431 | DeepSeek V4 Pro 0813 |
| LMArena Vision: diagrams | Not measured | 1265 | Only GPT-5.2 measured |
| LMArena Vision: homework | Not measured | 1276 | Only GPT-5.2 measured |
| Kagi LLM Benchmark | Not measured | 73.3% | Only GPT-5.2 measured |
| Neither model has a result on the other 4 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index42
| Benchmark | DeepSeek V4 Pro 0813 | GPT-5.2 | Leader |
|---|---|---|---|
| AIME | Not measured | 96.9% | Only GPT-5.2 measured |
| MGSM (Vals AI) | Not measured | 94.0% | Only GPT-5.2 measured |
| AA Math Index | Not measured | 99.0 | Only GPT-5.2 measured |
| AIME 2025 (Artificial Analysis run) | Not measured | 99.0% | Only GPT-5.2 measured |
| OTIS Mock AIME | 98.6% | 96.1% | DeepSeek V4 Pro 0813 |
| FrontierMath, tiers 1 to 3 | 64.6% | 67.4% | GPT-5.2 |
| FrontierMath, tier 4 | 26.8% | 31.7% | GPT-5.2 |
| ProofBench | 50.0% | 15.0% | DeepSeek V4 Pro 0813 |
| LMArena Text: math | 1441 | 1440 | DeepSeek V4 Pro 0813 |
| LMArena Text: mathematical work | 1443 | 1442 | DeepSeek V4 Pro 0813 |
| Neither model has a result on the other 3 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index64
| Benchmark | DeepSeek V4 Pro 0813 | GPT-5.2 | Leader |
|---|---|---|---|
| MMLU Pro | 87.0% | 86.2% | DeepSeek V4 Pro 0813 |
| LegalBench (Vals AI) | 82.4% | 82.8% | GPT-5.2 |
| CaseLaw (Vals AI) | Not measured | 66.0% | Only GPT-5.2 measured |
| CorpFin (Vals AI) | 65.4% | 65.9% | GPT-5.2 |
| TaxEval (Vals AI) | 73.1% | 75.8% | GPT-5.2 |
| MortgageTax (Vals AI) | Not measured | 67.1% | Only GPT-5.2 measured |
| MedQA (Vals AI) | Not measured | 94.1% | Only GPT-5.2 measured |
| MedCode (Vals AI) | 42.5% | 49.8% | GPT-5.2 |
| MMLU Pro (Artificial Analysis run) | Not measured | 87.4% | Only GPT-5.2 measured |
| SimpleQA Verified | 52.9% | 37.1% | DeepSeek V4 Pro 0813 |
| LMArena Text: science | 1457 | 1425 | DeepSeek V4 Pro 0813 |
| LMArena Text: medicine and healthcare | 1446 | 1425 | DeepSeek V4 Pro 0813 |
| LMArena Text: legal and government | 1459 | 1427 | DeepSeek V4 Pro 0813 |
| LMArena Text: business and finance | 1442 | 1414 | DeepSeek V4 Pro 0813 |
| FACTS Benchmark Suite | Not measured | 48.5% | Only GPT-5.2 measured |
| Neither model has a result on the other 3 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
| Benchmark | DeepSeek V4 Pro 0813 | GPT-5.2 | Leader |
|---|---|---|---|
| Design Arena: websites | Not measured | 1195 | Only GPT-5.2 measured |
| Design Arena: UI components | Not measured | 1190 | Only GPT-5.2 measured |
| Design Arena: game dev | Not measured | 1195 | Only GPT-5.2 measured |
| Design Arena: data viz | Not measured | 1202 | Only GPT-5.2 measured |
| Design Arena: 3D | Not measured | 1079 | Only GPT-5.2 measured |
| Design Arena: all code categories | Not measured | 1170 | Only GPT-5.2 measured |
| Design Arena: SVG | Not measured | 1139 | Only GPT-5.2 measured |
| Design Arena: ASCII art | Not measured | 1152 | Only GPT-5.2 measured |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index153
| Benchmark | DeepSeek V4 Pro 0813 | GPT-5.2 | Leader |
|---|---|---|---|
| Vals Index | 53.0% | Not measured | Only DeepSeek V4 Pro 0813 measured |
| SAGE (Vals AI) | Not measured | 49.3% | Only GPT-5.2 measured |
| MedScribe (Vals AI) | 80.2% | 84.4% | GPT-5.2 |
| IFBench | Not measured | 75.4% | Only GPT-5.2 measured |
| Long context reasoning (AA-LCR) | 80.3% | 82.7% | GPT-5.2 |
| DTBench | 93.9% | 90.9% | DeepSeek V4 Pro 0813 |
| LMCA | 45.5% | 43.9% | DeepSeek V4 Pro 0813 |
| CL-bench | Not measured | 18.2% | Only GPT-5.2 measured |
| Furniture assembly | Not measured | 38.3% | Only GPT-5.2 measured |
| LMArena Text | 1446 | 1416 | DeepSeek V4 Pro 0813 |
| LMArena Text: instruction following | 1448 | 1409 | DeepSeek V4 Pro 0813 |
| LMArena Text: creative writing | 1439 | 1376 | DeepSeek V4 Pro 0813 |
| LMArena Vision | Not measured | 1246 | Only GPT-5.2 measured |
| LMArena Text: English | 1457 | 1422 | DeepSeek V4 Pro 0813 |
| LMArena Text: non-English | 1431 | 1405 | DeepSeek V4 Pro 0813 |
| LMArena Text: Chinese | 1481 | 1447 | DeepSeek V4 Pro 0813 |
| LMArena Text: German | Not measured | 1444 | Only GPT-5.2 measured |
| LMArena Text: Japanese | Not measured | 1420 | Only GPT-5.2 measured |
| LMArena Text: Korean | 1449 | 1374 | DeepSeek V4 Pro 0813 |
| LMArena Text: French | 1446 | 1454 | GPT-5.2 |
| LMArena Text: Spanish | 1435 | 1413 | DeepSeek V4 Pro 0813 |
| LMArena Text: Russian | 1448 | 1415 | DeepSeek V4 Pro 0813 |
| LMArena Text: Polish | Not measured | 1426 | Only GPT-5.2 measured |
| LMArena Text: multi-turn | 1454 | 1419 | DeepSeek V4 Pro 0813 |
| LMArena Text: longer queries | 1458 | 1413 | DeepSeek V4 Pro 0813 |
| LMArena Text: writing and language | 1443 | 1392 | DeepSeek V4 Pro 0813 |
| LMArena Text: entertainment and media | 1424 | 1378 | DeepSeek V4 Pro 0813 |
| LMArena Vision: English | Not measured | 1243 | Only GPT-5.2 measured |
| LMArena Vision: Chinese | Not measured | 1305 | Only GPT-5.2 measured |
| LMArena Vision: reading text | Not measured | 1257 | Only GPT-5.2 measured |
| LMArena Vision: humor | Not measured | 1227 | Only GPT-5.2 measured |
| LMArena Vision: captioning | Not measured | 1267 | Only GPT-5.2 measured |
| LMArena Vision: recognizing things | Not measured | 1232 | Only GPT-5.2 measured |
| LMArena Vision: creative writing | Not measured | 1225 | Only GPT-5.2 measured |
| LMArena Document | Not measured | 1405 | Only GPT-5.2 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 Pro 0813 | GPT-5.2 | 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 Pro 0813 costs less
Repository review
50K input + 3K output tokens
DeepSeek V4 Pro 0813 costs less
Agent loop
200K input + 10K output tokens
DeepSeek V4 Pro 0813 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 Pro 0813 | OpenAI GPT-5.2 |
|---|---|---|
| Maker | DeepSeek | OpenAI |
| Weights | Open weights | No public weights listed |
| Released | August 12, 2026 | December 10, 2025 |
| Context window | 1.05M | 400K |
| Input $/Mtok | $0.66 | $1.75 |
| Output $/Mtok | $1.98 | $14.00 |
| Coding index | 56.1 | 55.9 |
| Independent measurements behind it | 5 | 3 |
| Cost per solved task | $0.10 | $1.91 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is DeepSeek V4 Pro 0813 better than GPT-5.2 for coding?
DeepSeek V4 Pro 0813 and GPT-5.2 are level for coding, 56.1 against 55.9 on the Vibe Coding Index, so price and context decide. Pick GPT-5.2 for work measured by WeirdML or Chess puzzles, where it leads. Pick DeepSeek V4 Pro 0813 for output tokens 7.1x cheaper ($1.98 vs $14.00 per million) and a lower cost per solved task ($0.10 against $1.91).
Which is cheaper, DeepSeek V4 Pro 0813 or GPT-5.2?
DeepSeek V4 Pro 0813 costs $0.66 per million input tokens and $1.98 per million output tokens; GPT-5.2 costs $1.75 and $14.00. Per solved SWE-bench Verified task, DeepSeek V4 Pro 0813 costs $0.10 and GPT-5.2 $1.91.
Which has the bigger context window, DeepSeek V4 Pro 0813 or GPT-5.2?
DeepSeek V4 Pro 0813, with 1.05M tokens against 400K.
How many benchmarks is the DeepSeek V4 Pro 0813 vs GPT-5.2 comparison based on?
56 benchmarks have a result for both models as of October 22, 2026. DeepSeek V4 Pro 0813's coding index rests on 5 independent benchmarks and GPT-5.2's on 3. Where only one model has a result, the page says so instead of naming a winner.