LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

DeepSeek V4 Pro 0423 vs GPT-5

Updated October 21, 2026. 64 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace DeepSeek V4 Pro 0423 Replace GPT-5
DeepSeek V4 Pro 0423DeepSeek
Coding
35.3
Overall
30.4
Output $/M
$1.9105
48
GPT-5OpenAI
Coding
72.2
Overall
23.0
Output $/M
$10.00
15
vs boards won of 64 shared, 1 tied

Decision reading

Pick GPT-5 for coding: it scores 72.2 on the Vibe Coding Index against 35.3 for DeepSeek V4 Pro 0423.

Pick DeepSeek V4 Pro 0423 when output tokens cost 5.2x less, or when the context window is larger, 1.05M against 400K tokens.

Model ADeepSeek V4 Pro 0423DeepSeek35.3/1001 independent benchmark$0.57 per solved taskModel BGPT-5OpenAI72.2/1003 independent benchmarks$0.39 per solved task
Shared results
64
DeepSeek V4 Pro 0423 only
19
GPT-5 only
45
Like-for-like categories
7 of 9

Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.

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.

At a glance

Coding index
DeepSeek V4 Pro 0423 35.3 from 1 benchmark, GPT-5 72.2 from 3 benchmarks.
Widest category
Coding, 9.9 points apart on average over 10 shared benchmarks.
Widest single gap
Vibe Code Bench (Vals AI): DeepSeek V4 Pro 0423 49.9% against GPT-5 20.1%, 29.8 points apart on a 0 to 100 scale.
Output price
DeepSeek V4 Pro 0423 $1.9105 against GPT-5 $10.00 per million output tokens, 5.2x apart.
Context window
DeepSeek V4 Pro 0423 takes 1.05M tokens, GPT-5 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.

Boards won, where both have a result
CategoryDeepSeek V4 Pro 0423BarsGPT-5Ties
Coding91
Intelligence11
Reasoning52
Math33
Knowledge65
Web design62
General1811 tie
All shared boards48151 tie
CodingLike-for-likeFour of these boards feed the Vibe Coding Index91
Coding benchmarks for DeepSeek V4 Pro 0423 and GPT-5
BenchmarkDeepSeek V4 Pro 0423GPT-5Leader
Aider polyglotNot measured88.0%Only GPT-5 measured
SWE-bench VerifiedNot measured74.4%Only GPT-5 measured
LMArena WebDev14471418DeepSeek V4 Pro 0423
AA Coding Index59.437.8DeepSeek V4 Pro 0423
LiveCodeBench (Vals AI)87.5%85.9%DeepSeek V4 Pro 0423
SWE-bench Verified (Vals AI)77.4%69.0%DeepSeek V4 Pro 0423
Terminal-Bench 4 (Vals AI)11.1%Not measuredOnly DeepSeek V4 Pro 0423 measured
Vibe Code Bench (Vals AI)49.9%20.1%DeepSeek V4 Pro 0423
Code migration (Vals AI)26.2%Not measuredOnly DeepSeek V4 Pro 0423 measured
ProgramBench (Vals AI)0.0%Not measuredOnly DeepSeek V4 Pro 0423 measured
Terminal-Bench 2.1 (Vals AI)50.2%Not measuredOnly DeepSeek V4 Pro 0423 measured
SkillsBench (Vals AI)51.3%Not measuredOnly DeepSeek V4 Pro 0423 measured
LiveCodeBench (Artificial Analysis run)Not measured84.6%Only GPT-5 measured
Terminal-Bench Hard (Artificial Analysis run)Not measured37.9%Only GPT-5 measured
Terminal-BenchNot measured49.6%Only GPT-5 measured
FrontierCode17.6%Not measuredOnly DeepSeek V4 Pro 0423 measured
GSO-BenchNot measured6.9%Only GPT-5 measured
WeirdML48.9%60.7%GPT-5
ALE-Bench1006.1807.6DeepSeek V4 Pro 0423
LMArena Text: coding14701436DeepSeek V4 Pro 0423
LMArena Text: software and IT14691420DeepSeek V4 Pro 0423
LMArena WebDev: React1441Not measuredOnly DeepSeek V4 Pro 0423 measured
LMArena WebDev: HTML14351427DeepSeek V4 Pro 0423
Neither model has a result on the other 15 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index11
Intelligence benchmarks for DeepSeek V4 Pro 0423 and GPT-5
BenchmarkDeepSeek V4 Pro 0423GPT-5Leader
AA Intelligence Index30.423.0DeepSeek V4 Pro 0423
Epoch Capabilities Index149.1150.0GPT-5
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index15 boards, none shared
Agentic work benchmarks for DeepSeek V4 Pro 0423 and GPT-5
BenchmarkDeepSeek V4 Pro 0423GPT-5Leader
AA Agentic Index26.3Not measuredOnly DeepSeek V4 Pro 0423 measured
Design Arena: agents building web apps1000Not measuredOnly DeepSeek V4 Pro 0423 measured
Design Arena: agents building full-stack apps948Not measuredOnly DeepSeek V4 Pro 0423 measured
Design Arena: agents building Godot games1059Not measuredOnly DeepSeek V4 Pro 0423 measured
Finance Agent (Vals AI)44.1%Not measuredOnly DeepSeek V4 Pro 0423 measured
Tax Agent Bench (Vals AI)25.2%Not measuredOnly DeepSeek V4 Pro 0423 measured
Legal Research Bench (Vals AI)23.1%Not measuredOnly DeepSeek V4 Pro 0423 measured
Excel Modeling Benchmark (Vals AI)51.6%Not measuredOnly DeepSeek V4 Pro 0423 measured
τ²-Bench TelecomNot measured86.5%Only GPT-5 measured
METR time horizonsNot measured69.6%Only GPT-5 measured
Remote Labor IndexNot measured1.7%Only GPT-5 measured
GDPvalNot measured34.8%Only GPT-5 measured
BALROGNot measured32.8%Only GPT-5 measured
DeepResearch BenchNot measured49.6%Only GPT-5 measured
Vending-Bench 2$3,285Not measuredOnly DeepSeek V4 Pro 0423 measured
Neither model has a result on the other 24 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index52
Reasoning benchmarks for DeepSeek V4 Pro 0423 and GPT-5
BenchmarkDeepSeek V4 Pro 0423GPT-5Leader
GPQA Diamond89.4%85.6%DeepSeek V4 Pro 0423
MMMU Pro (Vals AI)Not measured81.5%Only GPT-5 measured
GPQA Diamond (Artificial Analysis run)Not measured85.4%Only GPT-5 measured
Humanity's Last Exam (Artificial Analysis run)Not measured28.5%Only GPT-5 measured
GPQA Diamond (Epoch AI run)90.9%86.2%DeepSeek V4 Pro 0423
ARC-AGINot measured65.7%Only GPT-5 measured
ARC-AGI-2Not measured9.9%Only GPT-5 measured
SimpleBenchNot measured56.7%Only GPT-5 measured
Chess puzzles19.0%36.0%GPT-5
Mystery game puzzles17.0%23.0%GPT-5
EBR-benchNot measured12.7%Only GPT-5 measured
VPCTNot measured66.0%Only GPT-5 measured
LMArena Text: hard prompts14611415DeepSeek V4 Pro 0423
LMArena Text: expert14631419DeepSeek V4 Pro 0423
LMArena Text: hard prompts in English14661419DeepSeek V4 Pro 0423
LMArena Vision: diagramsNot measured1252Only GPT-5 measured
LMArena Vision: homeworkNot measured1278Only GPT-5 measured
Neither model has a result on the other 4 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index33
Math benchmarks for DeepSeek V4 Pro 0423 and GPT-5
BenchmarkDeepSeek V4 Pro 0423GPT-5Leader
AIMENot measured93.4%Only GPT-5 measured
MATH 500 (Vals AI)Not measured96.0%Only GPT-5 measured
MGSM (Vals AI)Not measured92.8%Only GPT-5 measured
AA Math IndexNot measured94.3Only GPT-5 measured
AIME 2025 (Artificial Analysis run)Not measured94.3%Only GPT-5 measured
OTIS Mock AIME96.7%91.4%DeepSeek V4 Pro 0423
FrontierMath, tiers 1 to 345.3%55.4%GPT-5
FrontierMath, tier 42.4%22.0%GPT-5
ProofBench16.0%18.0%GPT-5
MATH level 5Not measured98.1%Only GPT-5 measured
LMArena Text: math14431407DeepSeek V4 Pro 0423
LMArena Text: mathematical work14551405DeepSeek V4 Pro 0423
Neither model has a result on the other 1 board here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index65
Knowledge benchmarks for DeepSeek V4 Pro 0423 and GPT-5
BenchmarkDeepSeek V4 Pro 0423GPT-5Leader
MMLU Pro87.3%86.5%DeepSeek V4 Pro 0423
LegalBench (Vals AI)80.3%86.0%GPT-5
CaseLaw (Vals AI)59.4%66.5%GPT-5
CorpFin (Vals AI)61.4%61.1%DeepSeek V4 Pro 0423
TaxEval (Vals AI)72.1%73.4%GPT-5
MortgageTax (Vals AI)Not measured65.5%Only GPT-5 measured
MedQA (Vals AI)Not measured96.3%Only GPT-5 measured
MedCode (Vals AI)40.5%49.6%GPT-5
MMLU Pro (Artificial Analysis run)Not measured87.1%Only GPT-5 measured
SimpleQA Verified47.0%50.1%GPT-5
GeoBenchNot measured81.0%Only GPT-5 measured
LMArena Text: science14701416DeepSeek V4 Pro 0423
LMArena Text: medicine and healthcare14661413DeepSeek V4 Pro 0423
LMArena Text: legal and government14641431DeepSeek V4 Pro 0423
LMArena Text: business and finance14431400DeepSeek V4 Pro 0423
Neither model has a result on the other 2 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index62
Web design benchmarks for DeepSeek V4 Pro 0423 and GPT-5
BenchmarkDeepSeek V4 Pro 0423GPT-5Leader
Design Arena: websites12351185DeepSeek V4 Pro 0423
Design Arena: UI components12211179DeepSeek V4 Pro 0423
Design Arena: game dev12291185DeepSeek V4 Pro 0423
Design Arena: data viz12031221GPT-5
Design Arena: 3D12541056DeepSeek V4 Pro 0423
Design Arena: all code categories12401168DeepSeek V4 Pro 0423
Design Arena: SVG11391188GPT-5
Design Arena: ASCII art11481140DeepSeek 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 index181
General benchmarks for DeepSeek V4 Pro 0423 and GPT-5
BenchmarkDeepSeek V4 Pro 0423GPT-5Leader
Public Benefits Bench (Vals AI)62.9%Not measuredOnly DeepSeek V4 Pro 0423 measured
SAGE (Vals AI)Not measured43.7%Only GPT-5 measured
MedScribe (Vals AI)75.1%83.7%GPT-5
IFBenchNot measured73.1%Only GPT-5 measured
Long context reasoning (AA-LCR)Not measured78.2%Only GPT-5 measured
DTBench90.7%90.7%Tie
LMCA41.2%40.0%DeepSeek V4 Pro 0423
Fiction.LiveBenchNot measured97.2%Only GPT-5 measured
Creative writing (Lech Mazur)Not measured86.0%Only GPT-5 measured
Surface Evolver Bench40.0%Not measuredOnly DeepSeek V4 Pro 0423 measured
CL-bench Life13.5%Not measuredOnly DeepSeek V4 Pro 0423 measured
LMArena Text14511406DeepSeek V4 Pro 0423
LMArena Text: instruction following14451388DeepSeek V4 Pro 0423
LMArena Text: creative writing14461365DeepSeek V4 Pro 0423
LMArena VisionNot measured1232Only GPT-5 measured
LMArena Text: English14591409DeepSeek V4 Pro 0423
LMArena Text: non-English14391397DeepSeek V4 Pro 0423
LMArena Text: Chinese14811422DeepSeek V4 Pro 0423
LMArena Text: German14581416DeepSeek V4 Pro 0423
LMArena Text: Japanese14461409DeepSeek V4 Pro 0423
LMArena Text: Korean14091360DeepSeek V4 Pro 0423
LMArena Text: French14731409DeepSeek V4 Pro 0423
LMArena Text: Spanish14571399DeepSeek V4 Pro 0423
LMArena Text: Russian14531407DeepSeek V4 Pro 0423
LMArena Text: Polish14611431DeepSeek V4 Pro 0423
LMArena Text: multi-turn14681426DeepSeek V4 Pro 0423
LMArena Text: longer queries14581399DeepSeek V4 Pro 0423
LMArena Text: writing and language14471384DeepSeek V4 Pro 0423
LMArena Text: entertainment and media14311379DeepSeek V4 Pro 0423
LMArena Vision: EnglishNot measured1231Only GPT-5 measured
LMArena Vision: ChineseNot measured1267Only GPT-5 measured
LMArena Vision: reading textNot measured1246Only GPT-5 measured
LMArena Vision: humorNot measured1219Only GPT-5 measured
LMArena Vision: captioningNot measured1212Only GPT-5 measured
LMArena Vision: recognizing thingsNot measured1263Only GPT-5 measured
LMArena Vision: creative writingNot measured1210Only GPT-5 measured
Neither model has a result on the other 10 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for DeepSeek V4 Pro 0423 and GPT-5
BenchmarkDeepSeek V4 Pro 0423GPT-5Leader
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 0423$0.0019Fits in one request
GPT-5$0.0063Fits in one request

DeepSeek V4 Pro 0423 costs less

Repository review

50K input + 3K output tokens

DeepSeek V4 Pro 0423$0.053Fits in one request
GPT-5$0.092Fits in one request

DeepSeek V4 Pro 0423 costs less

Agent loop

200K input + 10K output tokens

DeepSeek V4 Pro 0423$0.210Fits in one request
GPT-5$0.350Fits in one request

DeepSeek V4 Pro 0423 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.

Specifications of DeepSeek V4 Pro 0423 and GPT-5
SpecificationDeepSeek V4 Pro 0423OpenAI GPT-5
MakerDeepSeekOpenAI
WeightsOpen weightsNo public weights listed
ReleasedApril 24, 2026August 7, 2025
Context window1.05M400K
Input $/Mtok$0.9553$1.25
Output $/Mtok$1.9105$10.00
Coding index35.372.2
Independent measurements behind it13
Cost per solved task$0.57$0.39
SourceOpenRouter listingOpenRouter listing

Questions

Is DeepSeek V4 Pro 0423 better than GPT-5 for coding?

Pick GPT-5 for coding: it scores 72.2 on the Vibe Coding Index against 35.3 for DeepSeek V4 Pro 0423. Pick DeepSeek V4 Pro 0423 when output tokens cost 5.2x less, or when the context window is larger, 1.05M against 400K tokens.

Which is cheaper, DeepSeek V4 Pro 0423 or GPT-5?

DeepSeek V4 Pro 0423 costs $0.9553 per million input tokens and $1.9105 per million output tokens; GPT-5 costs $1.25 and $10.00. Per solved SWE-bench Verified task, DeepSeek V4 Pro 0423 costs $0.57 and GPT-5 $0.39.

Which has the bigger context window, DeepSeek V4 Pro 0423 or GPT-5?

DeepSeek V4 Pro 0423, with 1.05M tokens against 400K.

How many benchmarks is the DeepSeek V4 Pro 0423 vs GPT-5 comparison based on?

64 benchmarks have a result for both models as of October 21, 2026. DeepSeek V4 Pro 0423's coding index rests on 1 independent benchmark and GPT-5's on 3. Where only one model has a result, the page says so instead of naming a winner.

Compare next