DeepSeek V4 Pro 0423 vs DeepSeek V4 Pro 0813

DeepSeek V4 Pro 0813 leads DeepSeek V4 Pro 0423 on coding, 56.1 to 38.7 on the Vibe Coding Index, and DeepSeek V4 Pro 0813 costs 1.4x less per output token ($1.98 vs $2.716 per million). Updated October 21, 2026; 64 benchmarks have a result for both.

Replace DeepSeek V4 Pro 0423 Replace DeepSeek V4 Pro 0813

DeepSeek V4 Pro 0423DeepSeek

Coding: 38.7
Overall: 50.0
Output $/M: $2.716

DeepSeek V4 Pro 0813DeepSeek

Coding: 56.1
Overall: 60.8
Output $/M: $1.98

vs boards won of 64 shared, 1 tied

DeepSeek V4 Pro 0423 or DeepSeek V4 Pro 0813: which to pick

Pick DeepSeek V4 Pro 0813 for coding: it scores 56.1 on the Vibe Coding Index against 38.7 for DeepSeek V4 Pro 0423.

Pick DeepSeek V4 Pro 0423 for work measured by LMArena Text: coding or LMArena Text: software and IT, where it leads.

Model ADeepSeek V4 Pro 0423DeepSeek38.7/1005 independent benchmarks$0.57 per solved task Model BDeepSeek V4 Pro 0813DeepSeek56.1/1005 independent benchmarks$0.10 per solved task

642014

Shared results: 64
DeepSeek V4 Pro 0423 only: 20
DeepSeek V4 Pro 0813 only: 14
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
DeepSeek V4 Pro 0813
DeepSeek V4 Pro 0813 scores 56.1 against 38.7.
5 benchmarks behind each score
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
DeepSeek V4 Pro 0813
DeepSeek V4 Pro 0813 leads, 41.3 against 26.3.
Same board, both measured
Reasoning
GPQA Diamond
DeepSeek V4 Pro 0813
DeepSeek V4 Pro 0813 leads, 92.4% against 89.4%.
Same board, both measured
Knowledge
MMLU Pro
DeepSeek V4 Pro 0423
DeepSeek V4 Pro 0423 leads, 87.3% against 87.0%.
Same board, both measured
Web design
Design Arena: websites
No call
Only DeepSeek V4 Pro 0423 has a Design Arena: websites result, 1234.
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
No call
Both take 1.05M tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
DeepSeek V4 Pro 0813
DeepSeek V4 Pro 0423 $0.0026, DeepSeek V4 Pro 0813 $0.0016 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
DeepSeek V4 Pro 0813
DeepSeek V4 Pro 0423 $0.57, DeepSeek V4 Pro 0813 $0.10.
Published attempt cost and pass rate

At a glance

Coding index: DeepSeek V4 Pro 0423 38.7 from 5 benchmarks, DeepSeek V4 Pro 0813 56.1 from 5 benchmarks.
Widest category: Math, 13.7 points apart on average over 6 shared benchmarks.
Widest single gap: ProofBench: DeepSeek V4 Pro 0423 16.0% against DeepSeek V4 Pro 0813 50.0%, 34.0 points apart on a 0 to 100 scale.
Output price: DeepSeek V4 Pro 0423 $2.716 against DeepSeek V4 Pro 0813 $1.98 per million output tokens, 1.4x 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.

Boards won, where both have a result
Category	DeepSeek V4 Pro 0423	DeepSeek V4 Pro 0813	Ties
Coding	2	14	1 tie
Intelligence	0	2
Agentic work	0	5
Reasoning	3	4
Math	2	4
Knowledge	5	5
General	12	5
All shared boards	24	39	1 tie

CodingLike-for-likeSix of these boards feed the Vibe Coding Index214

Coding benchmarks for DeepSeek V4 Pro 0423 and DeepSeek V4 Pro 0813
Benchmark	DeepSeek V4 Pro 0423	DeepSeek V4 Pro 0813	Leader
LMArena WebDev	1447	1582	DeepSeek V4 Pro 0813
AA Coding Index	59.4	68.8	DeepSeek V4 Pro 0813
LiveCodeBench (Vals AI)	87.5%	87.5%	DeepSeek V4 Pro 0813
SWE-bench Verified (Vals AI)	77.4%	96.4%	DeepSeek V4 Pro 0813
Terminal-Bench 4 (Vals AI)	11.1%	14.1%	DeepSeek V4 Pro 0813
Vibe Code Bench (Vals AI)	49.9%	82.3%	DeepSeek V4 Pro 0813
Code migration (Vals AI)	26.2%	41.5%	DeepSeek V4 Pro 0813
ProgramBench (Vals AI)	0.0%	0.0%	Tie
IOI (Vals AI)	Not measured	51.6%	Only DeepSeek V4 Pro 0813 measured
Terminal-Bench 2.1 (Vals AI)	50.2%	54.7%	DeepSeek V4 Pro 0813
Vibe Code Bench 1-100 (Vals AI)	Not measured	17.5%	Only DeepSeek V4 Pro 0813 measured
SkillsBench (Vals AI)	51.3%	53.8%	DeepSeek V4 Pro 0813
CyberBench (Vals AI)	Not measured	72.9%	Only DeepSeek V4 Pro 0813 measured
SciCode	Not measured	51.0%	Only DeepSeek V4 Pro 0813 measured
FrontierCode	17.6%	28.5%	DeepSeek V4 Pro 0813
WeirdML	48.9%	66.2%	DeepSeek V4 Pro 0813
ALE-Bench	1006.1	1403.2	DeepSeek V4 Pro 0813
LMArena Text: coding	1470	1469	DeepSeek V4 Pro 0423
LMArena Text: software and IT	1469	1462	DeepSeek V4 Pro 0423
LMArena WebDev: React	1441	1588	DeepSeek V4 Pro 0813
LMArena WebDev: HTML	1435	1566	DeepSeek V4 Pro 0813
Neither model has a result on the other 18 boards here

IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02

Intelligence benchmarks for DeepSeek V4 Pro 0423 and DeepSeek V4 Pro 0813
Benchmark	DeepSeek V4 Pro 0423	DeepSeek V4 Pro 0813	Leader
AA Intelligence Index	30.4	36.0	DeepSeek V4 Pro 0813
Epoch Capabilities Index	149.1	155.6	DeepSeek V4 Pro 0813

Agentic workLike-for-likePublic boards, shown as published, kept out of the index05

Agentic work benchmarks for DeepSeek V4 Pro 0423 and DeepSeek V4 Pro 0813
Benchmark	DeepSeek V4 Pro 0423	DeepSeek V4 Pro 0813	Leader
AA Agentic Index	26.3	41.3	DeepSeek V4 Pro 0813
Design Arena: agents building web apps	1000	Not measured	Only DeepSeek V4 Pro 0423 measured
Design Arena: agents building full-stack apps	948	Not measured	Only DeepSeek V4 Pro 0423 measured
Design Arena: agents building Godot games	1059	Not measured	Only DeepSeek V4 Pro 0423 measured
SREBench (Vals AI)	Not measured	1.9%	Only DeepSeek V4 Pro 0813 measured
Finance Agent (Vals AI)	44.1%	50.4%	DeepSeek V4 Pro 0813
Tax Agent Bench (Vals AI)	25.2%	26.8%	DeepSeek V4 Pro 0813
Legal Research Bench (Vals AI)	23.1%	40.9%	DeepSeek V4 Pro 0813
Harvey's Legal Agent Benchmark (Vals AI)	Not measured	10.8%	Only DeepSeek V4 Pro 0813 measured
Excel Modeling Benchmark (Vals AI)	51.6%	52.8%	DeepSeek V4 Pro 0813
Terminal-Bench-Science (Vals AI)	Not measured	4.3%	Only DeepSeek V4 Pro 0813 measured
APEX-Agents	Not measured	47.3%	Only DeepSeek V4 Pro 0813 measured
Vending-Bench 2	$3,285	Not measured	Only DeepSeek V4 Pro 0423 measured
Neither model has a result on the other 28 boards here

ReasoningLike-for-likePublic boards, shown as published, kept out of the index34

Reasoning benchmarks for DeepSeek V4 Pro 0423 and DeepSeek V4 Pro 0813
Benchmark	DeepSeek V4 Pro 0423	DeepSeek V4 Pro 0813	Leader
GPQA Diamond	89.4%	92.4%	DeepSeek V4 Pro 0813
GPQA Diamond (Artificial Analysis run)	Not measured	92.8%	Only DeepSeek V4 Pro 0813 measured
Humanity's Last Exam (Artificial Analysis run)	Not measured	41.0%	Only DeepSeek V4 Pro 0813 measured
GPQA Diamond (Epoch AI run)	90.9%	91.7%	DeepSeek V4 Pro 0813
ARC-AGI	Not measured	90.5%	Only DeepSeek V4 Pro 0813 measured
ARC-AGI-2	Not measured	61.3%	Only DeepSeek V4 Pro 0813 measured
Chess puzzles	19.0%	46.0%	DeepSeek V4 Pro 0813
Mystery game puzzles	17.0%	43.0%	DeepSeek V4 Pro 0813
LMArena Text: hard prompts	1461	1458	DeepSeek V4 Pro 0423
LMArena Text: expert	1463	1459	DeepSeek V4 Pro 0423
LMArena Text: hard prompts in English	1466	1463	DeepSeek V4 Pro 0423
Kagi LLM Benchmark	53.5%	Not measured	Only DeepSeek V4 Pro 0423 measured
Neither model has a result on the other 10 boards here

MathLike-for-likePublic boards, shown as published, kept out of the index24

Math benchmarks for DeepSeek V4 Pro 0423 and DeepSeek V4 Pro 0813
Benchmark	DeepSeek V4 Pro 0423	DeepSeek V4 Pro 0813	Leader
OTIS Mock AIME	96.7%	98.6%	DeepSeek V4 Pro 0813
FrontierMath, tiers 1 to 3	45.3%	64.6%	DeepSeek V4 Pro 0813
FrontierMath, tier 4	2.4%	26.8%	DeepSeek V4 Pro 0813
ProofBench	16.0%	50.0%	DeepSeek V4 Pro 0813
LMArena Text: math	1443	1441	DeepSeek V4 Pro 0423
LMArena Text: mathematical work	1455	1443	DeepSeek V4 Pro 0423
Neither model has a result on the other 7 boards here

KnowledgeLike-for-likePublic boards, shown as published, kept out of the index55

Knowledge benchmarks for DeepSeek V4 Pro 0423 and DeepSeek V4 Pro 0813
Benchmark	DeepSeek V4 Pro 0423	DeepSeek V4 Pro 0813	Leader
MMLU Pro	87.3%	87.0%	DeepSeek V4 Pro 0423
LegalBench (Vals AI)	80.3%	82.4%	DeepSeek V4 Pro 0813
CaseLaw (Vals AI)	59.4%	Not measured	Only DeepSeek V4 Pro 0423 measured
CorpFin (Vals AI)	61.4%	65.4%	DeepSeek V4 Pro 0813
TaxEval (Vals AI)	72.1%	73.1%	DeepSeek V4 Pro 0813
MedCode (Vals AI)	40.5%	42.5%	DeepSeek V4 Pro 0813
SimpleQA Verified	47.0%	52.9%	DeepSeek V4 Pro 0813
LMArena Text: science	1470	1457	DeepSeek V4 Pro 0423
LMArena Text: medicine and healthcare	1466	1446	DeepSeek V4 Pro 0423
LMArena Text: legal and government	1464	1459	DeepSeek V4 Pro 0423
LMArena Text: business and finance	1443	1442	DeepSeek V4 Pro 0423
Neither model has a result on the other 7 boards here

Web designDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared

Web design benchmarks for DeepSeek V4 Pro 0423 and DeepSeek V4 Pro 0813
Benchmark	DeepSeek V4 Pro 0423	DeepSeek V4 Pro 0813	Leader
Design Arena: websites	1234	Not measured	Only DeepSeek V4 Pro 0423 measured
Design Arena: UI components	1222	Not measured	Only DeepSeek V4 Pro 0423 measured
Design Arena: game dev	1229	Not measured	Only DeepSeek V4 Pro 0423 measured
Design Arena: data viz	1203	Not measured	Only DeepSeek V4 Pro 0423 measured
Design Arena: 3D	1254	Not measured	Only DeepSeek V4 Pro 0423 measured
Design Arena: all code categories	1240	Not measured	Only DeepSeek V4 Pro 0423 measured
Design Arena: SVG	1140	Not measured	Only DeepSeek V4 Pro 0423 measured
Design Arena: ASCII art	1149	Not measured	Only DeepSeek V4 Pro 0423 measured
Neither model has a result on the other 1 board here

GeneralLike-for-likePublic boards, shown as published, kept out of the index125

General benchmarks for DeepSeek V4 Pro 0423 and DeepSeek V4 Pro 0813
Benchmark	DeepSeek V4 Pro 0423	DeepSeek V4 Pro 0813	Leader
Vals Index	Not measured	53.0%	Only DeepSeek V4 Pro 0813 measured
Public Benefits Bench (Vals AI)	62.9%	Not measured	Only DeepSeek V4 Pro 0423 measured
MedScribe (Vals AI)	75.1%	80.2%	DeepSeek V4 Pro 0813
Long context reasoning (AA-LCR)	Not measured	80.3%	Only DeepSeek V4 Pro 0813 measured
DTBench	90.7%	93.9%	DeepSeek V4 Pro 0813
LMCA	41.2%	45.5%	DeepSeek V4 Pro 0813
Surface Evolver Bench	40.0%	Not measured	Only DeepSeek V4 Pro 0423 measured
CL-bench Life	13.5%	Not measured	Only DeepSeek V4 Pro 0423 measured
LMArena Text	1451	1446	DeepSeek V4 Pro 0423
LMArena Text: instruction following	1445	1448	DeepSeek V4 Pro 0813
LMArena Text: creative writing	1446	1439	DeepSeek V4 Pro 0423
LMArena Text: English	1459	1457	DeepSeek V4 Pro 0423
LMArena Text: non-English	1439	1431	DeepSeek V4 Pro 0423
LMArena Text: Chinese	1481	1481	DeepSeek V4 Pro 0423
LMArena Text: German	1458	Not measured	Only DeepSeek V4 Pro 0423 measured
LMArena Text: Japanese	1446	Not measured	Only DeepSeek V4 Pro 0423 measured
LMArena Text: Korean	1409	1449	DeepSeek V4 Pro 0813
LMArena Text: French	1473	1446	DeepSeek V4 Pro 0423
LMArena Text: Spanish	1457	1435	DeepSeek V4 Pro 0423
LMArena Text: Russian	1453	1448	DeepSeek V4 Pro 0423
LMArena Text: Polish	1461	Not measured	Only DeepSeek V4 Pro 0423 measured
LMArena Text: multi-turn	1468	1454	DeepSeek V4 Pro 0423
LMArena Text: longer queries	1458	1458	DeepSeek V4 Pro 0423
LMArena Text: writing and language	1447	1443	DeepSeek V4 Pro 0423
LMArena Text: entertainment and media	1431	1424	DeepSeek V4 Pro 0423
Neither model has a result on the other 21 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 DeepSeek V4 Pro 0813
Benchmark	DeepSeek V4 Pro 0423	DeepSeek V4 Pro 0813	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 0423$0.0026Fits in one request

DeepSeek V4 Pro 0813$0.0016Fits in one request

DeepSeek V4 Pro 0813 costs less

Repository review

50K input + 3K output tokens

DeepSeek V4 Pro 0423$0.069Fits in one request

DeepSeek V4 Pro 0813$0.039Fits in one request

DeepSeek V4 Pro 0813 costs less

Agent loop

200K input + 10K output tokens

DeepSeek V4 Pro 0423$0.272Fits in one request

DeepSeek V4 Pro 0813$0.152Fits in one request

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.

Specifications of DeepSeek V4 Pro 0423 and DeepSeek V4 Pro 0813
Specification	DeepSeek V4 Pro 0423	DeepSeek V4 Pro 0813
Maker	DeepSeek	DeepSeek
Weights	Open weights	Open weights
Released	April 24, 2026	August 12, 2026
Context window	1.05M	1.05M
Input $/Mtok	$1.225	$0.66
Output $/Mtok	$2.716	$1.98
Coding index	38.7	56.1
Independent measurements behind it	5	5
Cost per solved task	$0.57	$0.10
Source	OpenRouter listing	OpenRouter listing

Questions

Is DeepSeek V4 Pro 0423 better than DeepSeek V4 Pro 0813 for coding?

Pick DeepSeek V4 Pro 0813 for coding: it scores 56.1 on the Vibe Coding Index against 38.7 for DeepSeek V4 Pro 0423. Pick DeepSeek V4 Pro 0423 for work measured by LMArena Text: coding or LMArena Text: software and IT, where it leads.

Which is cheaper, DeepSeek V4 Pro 0423 or DeepSeek V4 Pro 0813?

DeepSeek V4 Pro 0423 costs $1.225 per million input tokens and $2.716 per million output tokens; DeepSeek V4 Pro 0813 costs $0.66 and $1.98. Per solved SWE-bench Verified task, DeepSeek V4 Pro 0423 costs $0.57 and DeepSeek V4 Pro 0813 $0.10.

Which has the bigger context window, DeepSeek V4 Pro 0423 or DeepSeek V4 Pro 0813?

Neither: both take 1.05M tokens.

How many benchmarks is the DeepSeek V4 Pro 0423 vs DeepSeek V4 Pro 0813 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 5 independent benchmarks and DeepSeek V4 Pro 0813's on 5. Where only one model has a result, the page says so instead of naming a winner.

Compare next

Embed this comparison on your site

Free under CC BY 4.0, while the source link stays visible. Preview the frame · Terms and other embeds

<iframe src="https://llmbenchmark.ai/embed/compare/deepseek-v4-pro-vs-deepseek-v4-pro-0813" title="DeepSeek V4 Pro 0423 vs DeepSeek V4 Pro 0813, LLMBenchmark.ai" width="100%" height="360" style="border:0;max-width:100%" loading="lazy"></iframe>
<p style="font:13px/1.4 sans-serif;margin:6px 0 0">Source: <a href="https://llmbenchmark.ai/compare/deepseek-v4-pro-vs-deepseek-v4-pro-0813">LLMBenchmark.ai, DeepSeek V4 Pro 0423 vs DeepSeek V4 Pro 0813</a></p>

DeepSeek V4 Pro 0423 or DeepSeek V4 Pro 0813: which to pick

Which one for your work

Coding

Intelligence

Agentic work

Reasoning

Knowledge

Web design

General

Long documents

Everyday cost

Cost per solved task

At a glance

Benchmark by benchmark

What each workload costs

Chat turn

Repository review

Agent loop

Specification differences

Questions

Compare next