Enter opens the highlighted result. Esc closes.

DeepSeek V4 Pro 0423 vs Grok 4.3

DeepSeek V4 Pro 0423 leads Grok 4.3 on coding, 38.7 to 37.4 on the Vibe Coding Index, at about the same output price ($2.976 vs $2.50 per million). Updated October 22, 2026; 77 benchmarks have a result for both.

Replace DeepSeek V4 Pro 0423 Replace Grok 4.3
DeepSeek V4 Pro 0423DeepSeek
Coding
38.7
Overall
50.0
Output $/M
$2.976
67
Grok 4.3xAI
Coding
37.4
Overall
53.1
Output $/M
$2.50
8
vs boards won of 77 shared, 2 tied

DeepSeek V4 Pro 0423 or Grok 4.3: which to pick

DeepSeek V4 Pro 0423 and Grok 4.3 are level for coding, 38.7 against 37.4 on the Vibe Coding Index, so price and context decide.

Pick Grok 4.3 for output tokens 1.2x cheaper ($2.50 vs $2.976 per million) and a lower cost per solved task ($0.31 against $0.57). Pick DeepSeek V4 Pro 0423 for a larger context window (1.05M against 1M tokens) and work measured by LMArena WebDev or AA Coding Index, where it leads.

Model ADeepSeek V4 Pro 0423DeepSeek38.7/1005 independent benchmarks$0.57 per solved taskModel BGrok 4.3xAI37.4/1003 independent benchmarks$0.31 per solved task
Shared results
77
DeepSeek V4 Pro 0423 only
7
Grok 4.3 only
28
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.

At a glance

Coding index
DeepSeek V4 Pro 0423 38.7 from 5 benchmarks, Grok 4.3 37.4 from 3 benchmarks.
Widest category
Agentic work, 15.4 points apart on average over 9 shared benchmarks.
Widest single gap
Excel Modeling Benchmark (Vals AI): DeepSeek V4 Pro 0423 51.6% against Grok 4.3 18.4%, 33.2 points apart on a 0 to 100 scale.
Output price
DeepSeek V4 Pro 0423 $2.976 against Grok 4.3 $2.50 per million output tokens, 1.2x apart.
Context window
DeepSeek V4 Pro 0423 takes 1.05M tokens, Grok 4.3 1M.

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 0423BarsGrok 4.3Ties
Coding1311 tie
Intelligence10
Agentic work72
Reasoning51
Math51
Knowledge83
Web design80
General2001 tie
All shared boards6782 ties
CodingLike-for-likeSix of these boards feed the Vibe Coding Index131
Coding benchmarks for DeepSeek V4 Pro 0423 and Grok 4.3
BenchmarkDeepSeek V4 Pro 0423Grok 4.3Leader
LMArena WebDev14471358DeepSeek V4 Pro 0423
AA Coding Index59.442.2DeepSeek V4 Pro 0423
LiveCodeBench (Vals AI)87.5%84.5%DeepSeek V4 Pro 0423
SWE-bench Verified (Vals AI)77.4%71.4%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%19.4%DeepSeek V4 Pro 0423
Code migration (Vals AI)26.2%6.8%DeepSeek V4 Pro 0423
ProgramBench (Vals AI)0.0%0.0%Tie
Terminal-Bench 2.1 (Vals AI)50.2%42.0%DeepSeek V4 Pro 0423
SkillsBench (Vals AI)51.3%40.6%DeepSeek V4 Pro 0423
SciCodeNot measured48.3%Only Grok 4.3 measured
Terminal-Bench Hard (Artificial Analysis run)Not measured37.9%Only Grok 4.3 measured
FrontierCode17.6%Not measuredOnly DeepSeek V4 Pro 0423 measured
WeirdML48.9%49.9%Grok 4.3
ALE-Bench1006.1944.2DeepSeek V4 Pro 0423
LMArena Text: coding14701415DeepSeek V4 Pro 0423
LMArena Text: software and IT14691416DeepSeek V4 Pro 0423
LMArena WebDev: React14411353DeepSeek V4 Pro 0423
LMArena WebDev: HTML14351367DeepSeek V4 Pro 0423
LMArena WebDev: from an imageNot measured1375Only Grok 4.3 measured
Neither model has a result on the other 19 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
Intelligence benchmarks for DeepSeek V4 Pro 0423 and Grok 4.3
BenchmarkDeepSeek V4 Pro 0423Grok 4.3Leader
AA Intelligence Index30.424.9DeepSeek V4 Pro 0423
Epoch Capabilities Index149.1Not measuredOnly DeepSeek V4 Pro 0423 measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index72
Agentic work benchmarks for DeepSeek V4 Pro 0423 and Grok 4.3
BenchmarkDeepSeek V4 Pro 0423Grok 4.3Leader
AA Agentic Index26.315.5DeepSeek V4 Pro 0423
Design Arena: agents building web apps10001126Grok 4.3
Design Arena: agents building full-stack apps9481006Grok 4.3
Design Arena: agents building mobile appsNot measured1058Only Grok 4.3 measured
Design Arena: agents building Android appsNot measured962Only Grok 4.3 measured
Design Arena: agents building Godot games1059914DeepSeek V4 Pro 0423
Design Arena: agents building gamesNot measured1007Only Grok 4.3 measured
Design Arena: HTML slidesNot measured959Only Grok 4.3 measured
Design Arena: PowerPoint slidesNot measured988Only Grok 4.3 measured
Design Arena: spreadsheetsNot measured879Only Grok 4.3 measured
Finance Agent (Vals AI)44.1%37.7%DeepSeek V4 Pro 0423
Tax Agent Bench (Vals AI)25.2%6.2%DeepSeek V4 Pro 0423
Legal Research Bench (Vals AI)23.1%15.4%DeepSeek V4 Pro 0423
Excel Modeling Benchmark (Vals AI)51.6%18.4%DeepSeek V4 Pro 0423
τ²-Bench TelecomNot measured97.7%Only Grok 4.3 measured
Vending-Bench 2$3,285$35DeepSeek V4 Pro 0423
LMArena SearchNot measured1166Only Grok 4.3 measured
Neither model has a result on the other 24 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index51
Reasoning benchmarks for DeepSeek V4 Pro 0423 and Grok 4.3
BenchmarkDeepSeek V4 Pro 0423Grok 4.3Leader
GPQA Diamond89.4%91.4%Grok 4.3
MMMU Pro (Vals AI)Not measured83.1%Only Grok 4.3 measured
GPQA Diamond (Artificial Analysis run)Not measured90.1%Only Grok 4.3 measured
Humanity's Last Exam (Artificial Analysis run)Not measured37.2%Only Grok 4.3 measured
GPQA Diamond (Epoch AI run)90.9%88.8%DeepSeek V4 Pro 0423
Chess puzzles19.0%18.0%DeepSeek V4 Pro 0423
Mystery game puzzles17.0%Not measuredOnly DeepSeek V4 Pro 0423 measured
LMArena Text: hard prompts14611396DeepSeek V4 Pro 0423
LMArena Text: expert14631384DeepSeek V4 Pro 0423
LMArena Text: hard prompts in English14661402DeepSeek V4 Pro 0423
LMArena Vision: diagramsNot measured1241Only Grok 4.3 measured
LMArena Vision: homeworkNot measured1241Only Grok 4.3 measured
Kagi LLM Benchmark53.5%Not measuredOnly 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 index51
Math benchmarks for DeepSeek V4 Pro 0423 and Grok 4.3
BenchmarkDeepSeek V4 Pro 0423Grok 4.3Leader
OTIS Mock AIME96.7%93.3%DeepSeek V4 Pro 0423
FrontierMath, tiers 1 to 345.3%42.8%DeepSeek V4 Pro 0423
FrontierMath, tier 42.4%14.6%Grok 4.3
ProofBench16.0%11.0%DeepSeek V4 Pro 0423
LMArena Text: math14431387DeepSeek V4 Pro 0423
LMArena Text: mathematical work14551396DeepSeek 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 index83
Knowledge benchmarks for DeepSeek V4 Pro 0423 and Grok 4.3
BenchmarkDeepSeek V4 Pro 0423Grok 4.3Leader
MMLU Pro87.3%85.8%DeepSeek V4 Pro 0423
LegalBench (Vals AI)80.3%84.5%Grok 4.3
CaseLaw (Vals AI)59.4%79.3%Grok 4.3
CorpFin (Vals AI)61.4%68.5%Grok 4.3
TaxEval (Vals AI)72.1%70.8%DeepSeek V4 Pro 0423
MortgageTax (Vals AI)Not measured48.3%Only Grok 4.3 measured
MedCode (Vals AI)40.5%38.1%DeepSeek V4 Pro 0423
SimpleQA Verified47.0%33.2%DeepSeek V4 Pro 0423
LMArena Text: science14701399DeepSeek V4 Pro 0423
LMArena Text: medicine and healthcare14661384DeepSeek V4 Pro 0423
LMArena Text: legal and government14641386DeepSeek V4 Pro 0423
LMArena Text: business and finance14431381DeepSeek V4 Pro 0423
Neither model has a result on the other 6 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index80
Web design benchmarks for DeepSeek V4 Pro 0423 and Grok 4.3
BenchmarkDeepSeek V4 Pro 0423Grok 4.3Leader
Design Arena: websites12341196DeepSeek V4 Pro 0423
Design Arena: UI components12191190DeepSeek V4 Pro 0423
Design Arena: game dev12281173DeepSeek V4 Pro 0423
Design Arena: data viz12031179DeepSeek V4 Pro 0423
Design Arena: 3D12531124DeepSeek V4 Pro 0423
Design Arena: all code categories12401186DeepSeek V4 Pro 0423
Design Arena: SVG11391080DeepSeek V4 Pro 0423
Design Arena: ASCII art11491125DeepSeek 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 index200
General benchmarks for DeepSeek V4 Pro 0423 and Grok 4.3
BenchmarkDeepSeek V4 Pro 0423Grok 4.3Leader
Public Benefits Bench (Vals AI)62.9%51.7%DeepSeek V4 Pro 0423
SAGE (Vals AI)Not measured19.7%Only Grok 4.3 measured
MedScribe (Vals AI)75.1%74.4%DeepSeek V4 Pro 0423
Vals Multimodal IndexNot measured43.3%Only Grok 4.3 measured
IFBenchNot measured83.3%Only Grok 4.3 measured
Long context reasoning (AA-LCR)Not measured75.0%Only Grok 4.3 measured
DTBench90.7%90.7%Tie
LMCA41.2%38.3%DeepSeek V4 Pro 0423
Surface Evolver Bench40.0%Not measuredOnly DeepSeek V4 Pro 0423 measured
CL-bench Life13.5%Not measuredOnly DeepSeek V4 Pro 0423 measured
LMArena Text14511397DeepSeek V4 Pro 0423
LMArena Text: instruction following14451366DeepSeek V4 Pro 0423
LMArena Text: creative writing14461380DeepSeek V4 Pro 0423
LMArena VisionNot measured1229Only Grok 4.3 measured
LMArena Text: English14591404DeepSeek V4 Pro 0423
LMArena Text: non-English14391385DeepSeek V4 Pro 0423
LMArena Text: Chinese14811422DeepSeek V4 Pro 0423
LMArena Text: German14581395DeepSeek V4 Pro 0423
LMArena Text: Japanese14461379DeepSeek V4 Pro 0423
LMArena Text: Korean14091354DeepSeek V4 Pro 0423
LMArena Text: French14731411DeepSeek V4 Pro 0423
LMArena Text: Spanish14571399DeepSeek V4 Pro 0423
LMArena Text: Russian14531399DeepSeek V4 Pro 0423
LMArena Text: Polish14611398DeepSeek V4 Pro 0423
LMArena Text: multi-turn14681406DeepSeek V4 Pro 0423
LMArena Text: longer queries14581393DeepSeek V4 Pro 0423
LMArena Text: writing and language14471385DeepSeek V4 Pro 0423
LMArena Text: entertainment and media14311370DeepSeek V4 Pro 0423
LMArena Vision: EnglishNot measured1235Only Grok 4.3 measured
LMArena Vision: ChineseNot measured1257Only Grok 4.3 measured
LMArena Vision: reading textNot measured1235Only Grok 4.3 measured
LMArena Vision: humorNot measured1222Only Grok 4.3 measured
LMArena Vision: recognizing thingsNot measured1217Only Grok 4.3 measured
LMArena Vision: creative writingNot measured1237Only Grok 4.3 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
Image generation benchmarks for DeepSeek V4 Pro 0423 and Grok 4.3
BenchmarkDeepSeek V4 Pro 0423Grok 4.3Leader
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$0.0028Fits in one request
Grok 4.3$0.0025Fits in one request

Grok 4.3 costs less

Repository review

50K input + 3K output tokens

DeepSeek V4 Pro 0423$0.077Fits in one request
Grok 4.3$0.070Fits in one request

Grok 4.3 costs less

Agent loop

200K input + 10K output tokens

DeepSeek V4 Pro 0423$0.301Fits in one request
Grok 4.3$0.275Fits in one request

Grok 4.3 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 Grok 4.3
SpecificationDeepSeek V4 Pro 0423SpaceXAI Grok 4.3
MakerDeepSeekxAI
WeightsOpen weightsNo public weights listed
ReleasedApril 24, 2026April 30, 2026
Context window1.05M1M
Input $/Mtok$1.358$1.25
Output $/Mtok$2.976$2.50
Coding index38.737.4
Independent measurements behind it53
Cost per solved task$0.57$0.31
SourceOpenRouter listingOpenRouter listing

Questions

Is DeepSeek V4 Pro 0423 better than Grok 4.3 for coding?

DeepSeek V4 Pro 0423 and Grok 4.3 are level for coding, 38.7 against 37.4 on the Vibe Coding Index, so price and context decide. Pick Grok 4.3 for output tokens 1.2x cheaper ($2.50 vs $2.976 per million) and a lower cost per solved task ($0.31 against $0.57). Pick DeepSeek V4 Pro 0423 for a larger context window (1.05M against 1M tokens) and work measured by LMArena WebDev or AA Coding Index, where it leads.

Which is cheaper, DeepSeek V4 Pro 0423 or Grok 4.3?

DeepSeek V4 Pro 0423 costs $1.358 per million input tokens and $2.976 per million output tokens; Grok 4.3 costs $1.25 and $2.50. Per solved SWE-bench Verified task, DeepSeek V4 Pro 0423 costs $0.57 and Grok 4.3 $0.31.

Which has the bigger context window, DeepSeek V4 Pro 0423 or Grok 4.3?

DeepSeek V4 Pro 0423, with 1.05M tokens against 1M.

How many benchmarks is the DeepSeek V4 Pro 0423 vs Grok 4.3 comparison based on?

77 benchmarks have a result for both models as of October 22, 2026. DeepSeek V4 Pro 0423's coding index rests on 5 independent benchmarks and Grok 4.3's on 3. 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-grok-4-3" title="DeepSeek V4 Pro 0423 vs Grok 4.3, 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-grok-4-3">LLMBenchmark.ai, DeepSeek V4 Pro 0423 vs Grok 4.3</a></p>