DeepSeek V4 Pro 0423 vs Gemini 2.5 Pro

DeepSeek V4 Pro 0423 leads Gemini 2.5 Pro on coding, 38.7 to 19.1 on the Vibe Coding Index, and DeepSeek V4 Pro 0423 costs 3.4x less per output token ($2.976 vs $10.00 per million). Updated October 22, 2026; 55 benchmarks have a result for both.

Replace DeepSeek V4 Pro 0423 Replace Gemini 2.5 Pro

DeepSeek V4 Pro 0423DeepSeek

Coding: 38.7
Overall: 50.0
Output $/M: $2.976

Gemini 2.5 ProGoogle

Coding: 19.1
Overall: 41.8
Output $/M: $10.00

vs boards won of 55 shared

DeepSeek V4 Pro 0423 or Gemini 2.5 Pro: which to pick

Pick DeepSeek V4 Pro 0423 for coding: it scores 38.7 on the Vibe Coding Index against 19.1 for Gemini 2.5 Pro.

Pick Gemini 2.5 Pro for work measured by WeirdML or Kagi LLM Benchmark, where it leads.

Model ADeepSeek V4 Pro 0423DeepSeek38.7/1005 independent benchmarks$0.57 per solved task Model BGemini 2.5 ProGoogle19.1/1003 independent benchmarks$0.64 per solved task

552935

Shared results: 55
DeepSeek V4 Pro 0423 only: 29
Gemini 2.5 Pro only: 35
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 Pro 0423
DeepSeek V4 Pro 0423 scores 38.7 against 19.1.
5 and 3 benchmarks behind the scores
Intelligence
AA Intelligence Index
DeepSeek V4 Pro 0423
DeepSeek V4 Pro 0423 leads, 30.4 against 16.1.
Same board, both measured
Agentic work
AA Agentic Index
DeepSeek V4 Pro 0423
DeepSeek V4 Pro 0423 leads, 26.3 against 1.6.
Same board, both measured
Reasoning
GPQA Diamond
No call
Only DeepSeek V4 Pro 0423 has a GPQA Diamond result, 89.4%.
Directional only
Knowledge
MMLU Pro
No call
Only DeepSeek V4 Pro 0423 has a MMLU Pro result, 87.3%.
Directional only
Web design
Design Arena: websites
DeepSeek V4 Pro 0423
DeepSeek V4 Pro 0423 leads, 1234 against 1167.
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 Pro 0423
DeepSeek V4 Pro 0423 $0.0028, Gemini 2.5 Pro $0.0063 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
DeepSeek V4 Pro 0423
DeepSeek V4 Pro 0423 $0.57, Gemini 2.5 Pro $0.64.
Published attempt cost and pass rate

At a glance

Coding index: DeepSeek V4 Pro 0423 38.7 from 5 benchmarks, Gemini 2.5 Pro 19.1 from 3 benchmarks.
Widest category: Coding, 21.7 points apart on average over 8 shared benchmarks.
Widest single gap: Vibe Code Bench (Vals AI): DeepSeek V4 Pro 0423 49.9% against Gemini 2.5 Pro 0.4%, 49.5 points apart on a 0 to 100 scale.
Output price: DeepSeek V4 Pro 0423 $2.976 against Gemini 2.5 Pro $10.00 per million output tokens, 3.4x 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	Gemini 2.5 Pro
Coding	7	1
Intelligence	2	0
Agentic work	2	0
Reasoning	4	1
Math	4	1
Knowledge	2	5
Web design	5	1
General	8	12
All shared boards	34	21

CodingLike-for-likeSix of these boards feed the Vibe Coding Index71

Coding benchmarks for DeepSeek V4 Pro 0423 and Gemini 2.5 Pro
Benchmark	DeepSeek V4 Pro 0423	Gemini 2.5 Pro	Leader
SWE-bench Verified	Not measured	53.6%	Only Gemini 2.5 Pro measured
LMArena WebDev	1447	1226	DeepSeek V4 Pro 0423
AA Coding Index	59.4	33.3	DeepSeek V4 Pro 0423
LiveCodeBench (Vals AI)	87.5%	Not measured	Only DeepSeek V4 Pro 0423 measured
SWE-bench Verified (Vals AI)	77.4%	54.4%	DeepSeek V4 Pro 0423
Terminal-Bench 4 (Vals AI)	11.1%	Not measured	Only DeepSeek V4 Pro 0423 measured
Vibe Code Bench (Vals AI)	49.9%	0.4%	DeepSeek V4 Pro 0423
Code migration (Vals AI)	26.2%	Not measured	Only DeepSeek V4 Pro 0423 measured
ProgramBench (Vals AI)	0.0%	Not measured	Only DeepSeek V4 Pro 0423 measured
Terminal-Bench 2.1 (Vals AI)	50.2%	Not measured	Only DeepSeek V4 Pro 0423 measured
SkillsBench (Vals AI)	51.3%	Not measured	Only DeepSeek V4 Pro 0423 measured
LiveCodeBench (Artificial Analysis run)	Not measured	80.1%	Only Gemini 2.5 Pro measured
SciCode	Not measured	46.3%	Only Gemini 2.5 Pro measured
Terminal-Bench Hard (Artificial Analysis run)	Not measured	26.5%	Only Gemini 2.5 Pro measured
Terminal-Bench	Not measured	32.6%	Only Gemini 2.5 Pro measured
FrontierCode	17.6%	Not measured	Only DeepSeek V4 Pro 0423 measured
GSO-Bench	Not measured	3.9%	Only Gemini 2.5 Pro measured
WeirdML	48.9%	54.0%	Gemini 2.5 Pro
ALE-Bench	1006.1	Not measured	Only DeepSeek V4 Pro 0423 measured
LMArena Text: coding	1470	1452	DeepSeek V4 Pro 0423
LMArena Text: software and IT	1469	1459	DeepSeek V4 Pro 0423
LMArena WebDev: React	1441	Not measured	Only DeepSeek V4 Pro 0423 measured
LMArena WebDev: HTML	1435	1235	DeepSeek V4 Pro 0423
LMArena WebDev: from an image	Not measured	1285	Only Gemini 2.5 Pro measured
Neither model has a result on the other 15 boards here

IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20

Intelligence benchmarks for DeepSeek V4 Pro 0423 and Gemini 2.5 Pro
Benchmark	DeepSeek V4 Pro 0423	Gemini 2.5 Pro	Leader
AA Intelligence Index	30.4	16.1	DeepSeek V4 Pro 0423
Epoch Capabilities Index	149.1	145.1	DeepSeek V4 Pro 0423

Agentic workLike-for-likePublic boards, shown as published, kept out of the index20

Agentic work benchmarks for DeepSeek V4 Pro 0423 and Gemini 2.5 Pro
Benchmark	DeepSeek V4 Pro 0423	Gemini 2.5 Pro	Leader
AA Agentic Index	26.3	1.6	DeepSeek V4 Pro 0423
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
Finance Agent (Vals AI)	44.1%	Not measured	Only DeepSeek V4 Pro 0423 measured
Tax Agent Bench (Vals AI)	25.2%	Not measured	Only DeepSeek V4 Pro 0423 measured
Legal Research Bench (Vals AI)	23.1%	Not measured	Only DeepSeek V4 Pro 0423 measured
Excel Modeling Benchmark (Vals AI)	51.6%	Not measured	Only DeepSeek V4 Pro 0423 measured
τ²-Bench Telecom	Not measured	54.1%	Only Gemini 2.5 Pro measured
GDPval	Not measured	23.3%	Only Gemini 2.5 Pro measured
DeepResearch Bench	Not measured	41.5%	Only Gemini 2.5 Pro measured
Vending-Bench 2	$3,285	$574	DeepSeek V4 Pro 0423
BrowseComp-Plus	Not measured	29.5%	Only Gemini 2.5 Pro measured
Neither model has a result on the other 28 boards here

ReasoningLike-for-likePublic boards, shown as published, kept out of the index41

Reasoning benchmarks for DeepSeek V4 Pro 0423 and Gemini 2.5 Pro
Benchmark	DeepSeek V4 Pro 0423	Gemini 2.5 Pro	Leader
GPQA Diamond	89.4%	Not measured	Only DeepSeek V4 Pro 0423 measured
GPQA Diamond (Artificial Analysis run)	Not measured	84.4%	Only Gemini 2.5 Pro measured
Humanity's Last Exam (Artificial Analysis run)	Not measured	22.5%	Only Gemini 2.5 Pro measured
GPQA Diamond (Epoch AI run)	90.9%	85.3%	DeepSeek V4 Pro 0423
ARC-AGI	Not measured	41.0%	Only Gemini 2.5 Pro measured
ARC-AGI-2	Not measured	4.9%	Only Gemini 2.5 Pro measured
Chess puzzles	19.0%	Not measured	Only DeepSeek V4 Pro 0423 measured
Mystery game puzzles	17.0%	Not measured	Only DeepSeek V4 Pro 0423 measured
LMArena Text: hard prompts	1461	1455	DeepSeek V4 Pro 0423
LMArena Text: expert	1463	1452	DeepSeek V4 Pro 0423
LMArena Text: hard prompts in English	1466	1453	DeepSeek V4 Pro 0423
LMArena Vision: diagrams	Not measured	1274	Only Gemini 2.5 Pro measured
LMArena Vision: homework	Not measured	1281	Only Gemini 2.5 Pro measured
Kagi LLM Benchmark	53.5%	70.3%	Gemini 2.5 Pro
Neither model has a result on the other 8 boards here

MathLike-for-likePublic boards, shown as published, kept out of the index41

Math benchmarks for DeepSeek V4 Pro 0423 and Gemini 2.5 Pro
Benchmark	DeepSeek V4 Pro 0423	Gemini 2.5 Pro	Leader
AA Math Index	Not measured	87.7	Only Gemini 2.5 Pro measured
AIME 2025 (Artificial Analysis run)	Not measured	87.7%	Only Gemini 2.5 Pro measured
OTIS Mock AIME	96.7%	84.7%	DeepSeek V4 Pro 0423
FrontierMath, tiers 1 to 3	45.3%	24.6%	DeepSeek V4 Pro 0423
FrontierMath, tier 4	2.4%	0.0%	DeepSeek V4 Pro 0423
ProofBench	16.0%	Not measured	Only DeepSeek V4 Pro 0423 measured
LMArena Text: math	1443	1450	Gemini 2.5 Pro
LMArena Text: mathematical work	1455	1454	DeepSeek V4 Pro 0423
Neither model has a result on the other 5 boards here

KnowledgeLike-for-likePublic boards, shown as published, kept out of the index25

Knowledge benchmarks for DeepSeek V4 Pro 0423 and Gemini 2.5 Pro
Benchmark	DeepSeek V4 Pro 0423	Gemini 2.5 Pro	Leader
MMLU Pro	87.3%	Not measured	Only DeepSeek V4 Pro 0423 measured
LegalBench (Vals AI)	80.3%	Not measured	Only DeepSeek V4 Pro 0423 measured
CaseLaw (Vals AI)	59.4%	63.9%	Gemini 2.5 Pro
CorpFin (Vals AI)	61.4%	60.8%	DeepSeek V4 Pro 0423
TaxEval (Vals AI)	72.1%	Not measured	Only DeepSeek V4 Pro 0423 measured
MortgageTax (Vals AI)	Not measured	68.9%	Only Gemini 2.5 Pro measured
MedCode (Vals AI)	40.5%	50.6%	Gemini 2.5 Pro
MMLU Pro (Artificial Analysis run)	Not measured	86.2%	Only Gemini 2.5 Pro measured
SimpleQA Verified	47.0%	Not measured	Only DeepSeek V4 Pro 0423 measured
LMArena Text: science	1470	1480	Gemini 2.5 Pro
LMArena Text: medicine and healthcare	1466	1476	Gemini 2.5 Pro
LMArena Text: legal and government	1464	1480	Gemini 2.5 Pro
LMArena Text: business and finance	1443	1439	DeepSeek V4 Pro 0423
FACTS Benchmark Suite	Not measured	61.5%	Only Gemini 2.5 Pro measured
Neither model has a result on the other 4 boards here

Web designLike-for-likePublic boards, shown as published, kept out of the index51

Web design benchmarks for DeepSeek V4 Pro 0423 and Gemini 2.5 Pro
Benchmark	DeepSeek V4 Pro 0423	Gemini 2.5 Pro	Leader
Design Arena: websites	1234	1167	DeepSeek V4 Pro 0423
Design Arena: UI components	1219	1137	DeepSeek V4 Pro 0423
Design Arena: game dev	1228	1111	DeepSeek V4 Pro 0423
Design Arena: data viz	1203	1222	Gemini 2.5 Pro
Design Arena: 3D	1253	1080	DeepSeek V4 Pro 0423
Design Arena: all code categories	1240	1152	DeepSeek V4 Pro 0423
Design Arena: SVG	1139	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 index812

General benchmarks for DeepSeek V4 Pro 0423 and Gemini 2.5 Pro
Benchmark	DeepSeek V4 Pro 0423	Gemini 2.5 Pro	Leader
Public Benefits Bench (Vals AI)	62.9%	Not measured	Only DeepSeek V4 Pro 0423 measured
SAGE (Vals AI)	Not measured	41.9%	Only Gemini 2.5 Pro measured
MedScribe (Vals AI)	75.1%	73.5%	DeepSeek V4 Pro 0423
IFBench	Not measured	48.7%	Only Gemini 2.5 Pro measured
Long context reasoning (AA-LCR)	Not measured	69.0%	Only Gemini 2.5 Pro measured
DTBench	90.7%	82.4%	DeepSeek V4 Pro 0423
LMCA	41.2%	34.8%	DeepSeek V4 Pro 0423
Creative writing (Lech Mazur)	Not measured	83.8%	Only Gemini 2.5 Pro measured
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	1458	Gemini 2.5 Pro
LMArena Text: instruction following	1445	1437	DeepSeek V4 Pro 0423
LMArena Text: creative writing	1446	1454	Gemini 2.5 Pro
LMArena Vision	Not measured	1263	Only Gemini 2.5 Pro measured
LMArena Text: English	1459	1458	DeepSeek V4 Pro 0423
LMArena Text: non-English	1439	1452	Gemini 2.5 Pro
LMArena Text: Chinese	1481	1507	Gemini 2.5 Pro
LMArena Text: German	1458	1487	Gemini 2.5 Pro
LMArena Text: Japanese	1446	1461	Gemini 2.5 Pro
LMArena Text: Korean	1409	1434	Gemini 2.5 Pro
LMArena Text: French	1473	1473	DeepSeek V4 Pro 0423
LMArena Text: Spanish	1457	1474	Gemini 2.5 Pro
LMArena Text: Russian	1453	1461	Gemini 2.5 Pro
LMArena Text: Polish	1461	1479	Gemini 2.5 Pro
LMArena Text: multi-turn	1468	1453	DeepSeek V4 Pro 0423
LMArena Text: longer queries	1458	1449	DeepSeek V4 Pro 0423
LMArena Text: writing and language	1447	1454	Gemini 2.5 Pro
LMArena Text: entertainment and media	1431	1441	Gemini 2.5 Pro
LMArena Vision: English	Not measured	1261	Only Gemini 2.5 Pro measured
LMArena Vision: Chinese	Not measured	1300	Only Gemini 2.5 Pro measured
LMArena Vision: reading text	Not measured	1270	Only Gemini 2.5 Pro measured
LMArena Vision: humor	Not measured	1262	Only Gemini 2.5 Pro measured
LMArena Vision: captioning	Not measured	1304	Only Gemini 2.5 Pro measured
LMArena Vision: recognizing things	Not measured	1283	Only Gemini 2.5 Pro measured
LMArena Vision: creative writing	Not measured	1265	Only Gemini 2.5 Pro measured
LMArena Document	Not measured	1421	Only Gemini 2.5 Pro 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 Gemini 2.5 Pro
Benchmark	DeepSeek V4 Pro 0423	Gemini 2.5 Pro	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$0.0028Fits in one request

Gemini 2.5 Pro$0.0063Fits in one request

DeepSeek V4 Pro 0423 costs less

Repository review

50K input + 3K output tokens

DeepSeek V4 Pro 0423$0.077Fits in one request

Gemini 2.5 Pro$0.092Fits in one request

DeepSeek V4 Pro 0423 costs less

Agent loop

200K input + 10K output tokens

DeepSeek V4 Pro 0423$0.301Fits in one request

Gemini 2.5 Pro$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 Gemini 2.5 Pro
Specification	DeepSeek V4 Pro 0423	Google Gemini 2.5 Pro
Maker	DeepSeek	Google
Weights	Open weights	No public weights listed
Released	April 24, 2026	June 17, 2025
Context window	1.05M	1.05M
Input $/Mtok	$1.358	$1.25
Output $/Mtok	$2.976	$10.00
Coding index	38.7	19.1
Independent measurements behind it	5	3
Cost per solved task	$0.57	$0.64
Source	OpenRouter listing	OpenRouter listing

Questions

Is DeepSeek V4 Pro 0423 better than Gemini 2.5 Pro for coding?

Pick DeepSeek V4 Pro 0423 for coding: it scores 38.7 on the Vibe Coding Index against 19.1 for Gemini 2.5 Pro. Pick Gemini 2.5 Pro for work measured by WeirdML or Kagi LLM Benchmark, where it leads.

Which is cheaper, DeepSeek V4 Pro 0423 or Gemini 2.5 Pro?

DeepSeek V4 Pro 0423 costs $1.358 per million input tokens and $2.976 per million output tokens; Gemini 2.5 Pro costs $1.25 and $10.00. Per solved SWE-bench Verified task, DeepSeek V4 Pro 0423 costs $0.57 and Gemini 2.5 Pro $0.64.

Which has the bigger context window, DeepSeek V4 Pro 0423 or Gemini 2.5 Pro?

Neither: both take 1.05M tokens.

How many benchmarks is the DeepSeek V4 Pro 0423 vs Gemini 2.5 Pro comparison based on?

55 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 Gemini 2.5 Pro'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-gemini-2-5-pro" title="DeepSeek V4 Pro 0423 vs Gemini 2.5 Pro, 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-gemini-2-5-pro">LLMBenchmark.ai, DeepSeek V4 Pro 0423 vs Gemini 2.5 Pro</a></p>

DeepSeek V4 Pro 0423 or Gemini 2.5 Pro: which to pick

Which one for your work

Coding

Intelligence

Agentic work

Reasoning

Knowledge

Web design

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