DeepSeek V4 Flash 0731 vs GPT-5.4

GPT-5.4 leads DeepSeek V4 Flash 0731 on coding, 64.4 to 51.8 on the Vibe Coding Index, and DeepSeek V4 Flash 0731 costs 53.6x less per output token ($0.28 vs $15.00 per million). Updated October 21, 2026; 41 benchmarks have a result for both.

Replace DeepSeek V4 Flash 0731 Replace GPT-5.4

DeepSeek V4 Flash 0731DeepSeek

Coding: 51.8
Overall: 55.2
Output $/M: $0.28

GPT-5.4OpenAI

Coding: 64.4
Overall: 69.6
Output $/M: $15.00

vs boards won of 41 shared, 1 tied

DeepSeek V4 Flash 0731 or GPT-5.4: which to pick

Pick GPT-5.4 for coding: it scores 64.4 on the Vibe Coding Index against 51.8 for DeepSeek V4 Flash 0731, though on 3 benchmarks against 4.

Pick DeepSeek V4 Flash 0731 for output tokens 53.6x cheaper ($0.28 vs $15.00 per million) and a lower cost per solved task ($0.01 against $1.02).

Model ADeepSeek V4 Flash 0731DeepSeek51.8/1004 independent benchmarks$0.01 per solved task Model BGPT-5.4OpenAI64.4/1003 independent benchmarks$1.02 per solved task

411469

Shared results: 41
DeepSeek V4 Flash 0731 only: 14
GPT-5.4 only: 69
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
GPT-5.4
GPT-5.4 scores 64.4 against 51.8.
4 and 3 benchmarks behind the scores
Intelligence
AA Intelligence Index
GPT-5.4
GPT-5.4 leads, 39.0 against 34.3.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only DeepSeek V4 Flash 0731 has a AA Agentic Index result, 41.0.
Directional only
Reasoning
GPQA Diamond
GPT-5.4
GPT-5.4 leads, 91.7% against 89.9%.
Same board, both measured
Math
AIME
No call
Only GPT-5.4 has a AIME result, 96.7%.
Directional only
Knowledge
MMLU Pro
GPT-5.4
GPT-5.4 leads, 87.5% against 86.2%.
Same board, both measured
Web design
Design Arena: websites
DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 leads, 1238 against 1199.
Same board, both measured
Long documents
Listed context window
GPT-5.4
GPT-5.4 takes 1.05M tokens against 1.05M.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 $0.0003, GPT-5.4 $0.010 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 $0.01, GPT-5.4 $1.02.
Published attempt cost and pass rate

At a glance

Coding index: DeepSeek V4 Flash 0731 51.8 from 4 benchmarks, GPT-5.4 64.4 from 3 benchmarks.
Widest category: Math, 12.2 points apart on average over 4 shared benchmarks.
Widest single gap: FrontierMath, tier 4: DeepSeek V4 Flash 0731 24.4% against GPT-5.4 49.0%, 24.6 points apart on a 0 to 100 scale.
Output price: DeepSeek V4 Flash 0731 $0.28 against GPT-5.4 $15.00 per million output tokens, 53.6x apart.
Context window: DeepSeek V4 Flash 0731 takes 1.05M tokens, GPT-5.4 1.05M.

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 Flash 0731	GPT-5.4	Ties
Coding	4	5
Intelligence	0	2
Reasoning	0	8
Math	0	3	1 tie
Knowledge	1	5
Web design	6	2
General	1	3
All shared boards	12	28	1 tie

CodingLike-for-likeSix of these boards feed the Vibe Coding Index45

Coding benchmarks for DeepSeek V4 Flash 0731 and GPT-5.4
Benchmark	DeepSeek V4 Flash 0731	GPT-5.4	Leader
LMArena WebDev	Not measured	1464	Only GPT-5.4 measured
AA Coding Index	69.1	71.1	GPT-5.4
LiveCodeBench (Vals AI)	87.3%	84.1%	DeepSeek V4 Flash 0731
SWE-bench Verified (Vals AI)	88.8%	78.2%	DeepSeek V4 Flash 0731
Terminal-Bench 4 (Vals AI)	18.7%	Not measured	Only DeepSeek V4 Flash 0731 measured
Vibe Code Bench (Vals AI)	74.7%	67.4%	DeepSeek V4 Flash 0731
Code migration (Vals AI)	38.6%	35.0%	DeepSeek V4 Flash 0731
ProgramBench (Vals AI)	0.0%	0.5%	GPT-5.4
IOI (Vals AI)	32.7%	Not measured	Only DeepSeek V4 Flash 0731 measured
Terminal-Bench 2.1 (Vals AI)	67.0%	Not measured	Only DeepSeek V4 Flash 0731 measured
SkillsBench (Vals AI)	50.7%	51.7%	GPT-5.4
CyberBench (Vals AI)	72.1%	Not measured	Only DeepSeek V4 Flash 0731 measured
SciCode	50.3%	Not measured	Only DeepSeek V4 Flash 0731 measured
Terminal-Bench Hard (Artificial Analysis run)	Not measured	57.6%	Only GPT-5.4 measured
Terminal-Bench	Not measured	81.8%	Only GPT-5.4 measured
FrontierCode	18.8%	Not measured	Only DeepSeek V4 Flash 0731 measured
GSO-Bench	Not measured	31.4%	Only GPT-5.4 measured
WeirdML	63.0%	77.7%	GPT-5.4
MirrorCode	Not measured	15.6%	Only GPT-5.4 measured
ALE-Bench	1306.1	1607.0	GPT-5.4
LMArena Text: coding	Not measured	1496	Only GPT-5.4 measured
LMArena Text: software and IT	Not measured	1491	Only GPT-5.4 measured
LMArena WebDev: React	Not measured	1454	Only GPT-5.4 measured
LMArena WebDev: HTML	Not measured	1478	Only GPT-5.4 measured
LMArena WebDev: from an image	Not measured	1436	Only GPT-5.4 measured
Neither model has a result on the other 14 boards here

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

Intelligence benchmarks for DeepSeek V4 Flash 0731 and GPT-5.4
Benchmark	DeepSeek V4 Flash 0731	GPT-5.4	Leader
AA Intelligence Index	34.3	39.0	GPT-5.4
Epoch Capabilities Index	154.8	157.3	GPT-5.4

Agentic workDirectional onlyPublic boards, shown as published, kept out of the index18 boards, none shared

Agentic work benchmarks for DeepSeek V4 Flash 0731 and GPT-5.4
Benchmark	DeepSeek V4 Flash 0731	GPT-5.4	Leader
AA Agentic Index	41.0	Not measured	Only DeepSeek V4 Flash 0731 measured
Design Arena: agents building web apps	Not measured	1059	Only GPT-5.4 measured
Design Arena: agents building full-stack apps	Not measured	1017	Only GPT-5.4 measured
Design Arena: agents building mobile apps	Not measured	1087	Only GPT-5.4 measured
Design Arena: agents building Android apps	Not measured	1030	Only GPT-5.4 measured
Design Arena: agents building Godot games	Not measured	1135	Only GPT-5.4 measured
Finance Agent (Vals AI)	49.5%	Not measured	Only DeepSeek V4 Flash 0731 measured
Tax Agent Bench (Vals AI)	28.1%	Not measured	Only DeepSeek V4 Flash 0731 measured
Legal Research Bench (Vals AI)	30.3%	Not measured	Only DeepSeek V4 Flash 0731 measured
Excel Modeling Benchmark (Vals AI)	57.0%	Not measured	Only DeepSeek V4 Flash 0731 measured
BioMysteryBench (Vals AI)	64.4%	Not measured	Only DeepSeek V4 Flash 0731 measured
Terminal-Bench-Science (Vals AI)	4.3%	Not measured	Only DeepSeek V4 Flash 0731 measured
τ²-Bench Telecom	Not measured	87.1%	Only GPT-5.4 measured
APEX-Agents	Not measured	52.4%	Only GPT-5.4 measured
METR time horizons	Not measured	74.3%	Only GPT-5.4 measured
DeepResearch Bench	Not measured	35.1%	Only GPT-5.4 measured
PostTrainBench	Not measured	19.0%	Only GPT-5.4 measured
Vending-Bench 2	Not measured	$6,144	Only GPT-5.4 measured
Neither model has a result on the other 23 boards here

ReasoningLike-for-likePublic boards, shown as published, kept out of the index08

Reasoning benchmarks for DeepSeek V4 Flash 0731 and GPT-5.4
Benchmark	DeepSeek V4 Flash 0731	GPT-5.4	Leader
GPQA Diamond	89.9%	91.7%	GPT-5.4
MMMU Pro (Vals AI)	Not measured	87.5%	Only GPT-5.4 measured
GPQA Diamond (Artificial Analysis run)	90.8%	92.0%	GPT-5.4
Humanity's Last Exam (Artificial Analysis run)	38.6%	43.7%	GPT-5.4
GPQA Diamond (Epoch AI run)	91.0%	93.3%	GPT-5.4
ARC-AGI	89.0%	93.7%	GPT-5.4
ARC-AGI-2	61.4%	74.0%	GPT-5.4
SimpleBench	61.1%	Not measured	Only DeepSeek V4 Flash 0731 measured
Chess puzzles	32.0%	48.0%	GPT-5.4
Mystery game puzzles	34.0%	37.0%	GPT-5.4
EBR-bench	Not measured	25.4%	Only GPT-5.4 measured
LMArena Text: hard prompts	Not measured	1485	Only GPT-5.4 measured
LMArena Text: expert	Not measured	1507	Only GPT-5.4 measured
LMArena Text: hard prompts in English	Not measured	1479	Only GPT-5.4 measured
LMArena Vision: diagrams	Not measured	1327	Only GPT-5.4 measured
LMArena Vision: homework	Not measured	1336	Only GPT-5.4 measured
Kagi LLM Benchmark	Not measured	63.8%	Only GPT-5.4 measured
Neither model has a result on the other 5 boards here

MathLike-for-likePublic boards, shown as published, kept out of the index03

Math benchmarks for DeepSeek V4 Flash 0731 and GPT-5.4
Benchmark	DeepSeek V4 Flash 0731	GPT-5.4	Leader
AIME	Not measured	96.7%	Only GPT-5.4 measured
OTIS Mock AIME	94.4%	97.8%	GPT-5.4
FrontierMath, tiers 1 to 3	57.5%	78.6%	GPT-5.4
FrontierMath, tier 4	24.4%	49.0%	GPT-5.4
ProofBench	56.0%	56.0%	Tie
LMArena Text: math	Not measured	1488	Only GPT-5.4 measured
LMArena Text: mathematical work	Not measured	1497	Only GPT-5.4 measured
Neither model has a result on the other 6 boards here

KnowledgeLike-for-likePublic boards, shown as published, kept out of the index15

Knowledge benchmarks for DeepSeek V4 Flash 0731 and GPT-5.4
Benchmark	DeepSeek V4 Flash 0731	GPT-5.4	Leader
MMLU Pro	86.2%	87.5%	GPT-5.4
LegalBench (Vals AI)	77.7%	86.0%	GPT-5.4
CaseLaw (Vals AI)	Not measured	63.8%	Only GPT-5.4 measured
CorpFin (Vals AI)	61.9%	65.3%	GPT-5.4
TaxEval (Vals AI)	70.7%	74.0%	GPT-5.4
MortgageTax (Vals AI)	Not measured	68.3%	Only GPT-5.4 measured
MedQA (Vals AI)	Not measured	96.1%	Only GPT-5.4 measured
MedCode (Vals AI)	41.4%	41.3%	DeepSeek V4 Flash 0731
SimpleQA Verified	33.6%	45.1%	GPT-5.4
LMArena Text: science	Not measured	1479	Only GPT-5.4 measured
LMArena Text: medicine and healthcare	Not measured	1458	Only GPT-5.4 measured
LMArena Text: legal and government	Not measured	1486	Only GPT-5.4 measured
LMArena Text: business and finance	Not measured	1474	Only GPT-5.4 measured
Neither model has a result on the other 5 boards here

Web designLike-for-likePublic boards, shown as published, kept out of the index62

Web design benchmarks for DeepSeek V4 Flash 0731 and GPT-5.4
Benchmark	DeepSeek V4 Flash 0731	GPT-5.4	Leader
Design Arena: websites	1238	1199	DeepSeek V4 Flash 0731
Design Arena: UI components	1235	1215	DeepSeek V4 Flash 0731
Design Arena: game dev	1208	1195	DeepSeek V4 Flash 0731
Design Arena: data viz	1180	1196	GPT-5.4
Design Arena: 3D	1209	1066	DeepSeek V4 Flash 0731
Design Arena: all code categories	1230	1180	DeepSeek V4 Flash 0731
Design Arena: SVG	1185	1142	DeepSeek V4 Flash 0731
Design Arena: ASCII art	1090	1188	GPT-5.4
Neither model has a result on the other 1 board here

GeneralLike-for-likePublic boards, shown as published, kept out of the index13

General benchmarks for DeepSeek V4 Flash 0731 and GPT-5.4
Benchmark	DeepSeek V4 Flash 0731	GPT-5.4	Leader
SAGE (Vals AI)	Not measured	43.3%	Only GPT-5.4 measured
MedScribe (Vals AI)	80.4%	77.5%	DeepSeek V4 Flash 0731
IFBench	Not measured	73.9%	Only GPT-5.4 measured
Long context reasoning (AA-LCR)	79.7%	82.0%	GPT-5.4
DTBench	90.9%	94.4%	GPT-5.4
LMCA	41.7%	52.0%	GPT-5.4
CL-bench	Not measured	27.9%	Only GPT-5.4 measured
Furniture assembly	Not measured	37.5%	Only GPT-5.4 measured
CL-bench Life	Not measured	21.7%	Only GPT-5.4 measured
LMArena Text	Not measured	1469	Only GPT-5.4 measured
LMArena Text: instruction following	Not measured	1469	Only GPT-5.4 measured
LMArena Text: creative writing	Not measured	1439	Only GPT-5.4 measured
LMArena Vision	Not measured	1303	Only GPT-5.4 measured
LMArena Text: English	Not measured	1467	Only GPT-5.4 measured
LMArena Text: non-English	Not measured	1465	Only GPT-5.4 measured
LMArena Text: Chinese	Not measured	1520	Only GPT-5.4 measured
LMArena Text: German	Not measured	1472	Only GPT-5.4 measured
LMArena Text: Japanese	Not measured	1485	Only GPT-5.4 measured
LMArena Text: Korean	Not measured	1449	Only GPT-5.4 measured
LMArena Text: French	Not measured	1493	Only GPT-5.4 measured
LMArena Text: Spanish	Not measured	1452	Only GPT-5.4 measured
LMArena Text: Russian	Not measured	1480	Only GPT-5.4 measured
LMArena Text: Polish	Not measured	1472	Only GPT-5.4 measured
LMArena Text: multi-turn	Not measured	1482	Only GPT-5.4 measured
LMArena Text: longer queries	Not measured	1473	Only GPT-5.4 measured
LMArena Text: writing and language	Not measured	1458	Only GPT-5.4 measured
LMArena Text: entertainment and media	Not measured	1438	Only GPT-5.4 measured
LMArena Vision: English	Not measured	1301	Only GPT-5.4 measured
LMArena Vision: Chinese	Not measured	1354	Only GPT-5.4 measured
LMArena Vision: reading text	Not measured	1312	Only GPT-5.4 measured
LMArena Vision: humor	Not measured	1284	Only GPT-5.4 measured
LMArena Vision: recognizing things	Not measured	1251	Only GPT-5.4 measured
LMArena Vision: creative writing	Not measured	1283	Only GPT-5.4 measured
LMArena Document	Not measured	1471	Only GPT-5.4 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 Flash 0731 and GPT-5.4
Benchmark	DeepSeek V4 Flash 0731	GPT-5.4	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 Flash 0731$0.0003Fits in one request

GPT-5.4$0.010Fits in one request

DeepSeek V4 Flash 0731 costs less

Repository review

50K input + 3K output tokens

DeepSeek V4 Flash 0731$0.0078Fits in one request

GPT-5.4$0.170Fits in one request

DeepSeek V4 Flash 0731 costs less

Agent loop

200K input + 10K output tokens

DeepSeek V4 Flash 0731$0.031Fits in one request

GPT-5.4$0.650Fits in one request

DeepSeek V4 Flash 0731 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 Flash 0731 and GPT-5.4
Specification	DeepSeek V4 Flash 0731	OpenAI GPT-5.4
Maker	DeepSeek	OpenAI
Weights	Open weights	No public weights listed
Released	July 31, 2026	March 5, 2026
Context window	1.05M	1.05M
Input $/Mtok	$0.14	$2.50
Output $/Mtok	$0.28	$15.00
Coding index	51.8	64.4
Independent measurements behind it	4	3
Cost per solved task	$0.01	$1.02
Source	OpenRouter listing	OpenRouter listing

Questions

Is DeepSeek V4 Flash 0731 better than GPT-5.4 for coding?

Pick GPT-5.4 for coding: it scores 64.4 on the Vibe Coding Index against 51.8 for DeepSeek V4 Flash 0731, though on 3 benchmarks against 4. Pick DeepSeek V4 Flash 0731 for output tokens 53.6x cheaper ($0.28 vs $15.00 per million) and a lower cost per solved task ($0.01 against $1.02).

Which is cheaper, DeepSeek V4 Flash 0731 or GPT-5.4?

DeepSeek V4 Flash 0731 costs $0.14 per million input tokens and $0.28 per million output tokens; GPT-5.4 costs $2.50 and $15.00. Per solved SWE-bench Verified task, DeepSeek V4 Flash 0731 costs $0.01 and GPT-5.4 $1.02.

Which has the bigger context window, DeepSeek V4 Flash 0731 or GPT-5.4?

GPT-5.4, with 1.05M tokens against 1.05M.

How many benchmarks is the DeepSeek V4 Flash 0731 vs GPT-5.4 comparison based on?

41 benchmarks have a result for both models as of October 21, 2026. DeepSeek V4 Flash 0731's coding index rests on 4 independent benchmarks and GPT-5.4'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-flash-0731-vs-gpt-5-4" title="DeepSeek V4 Flash 0731 vs GPT-5.4, 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-flash-0731-vs-gpt-5-4">LLMBenchmark.ai, DeepSeek V4 Flash 0731 vs GPT-5.4</a></p>

DeepSeek V4 Flash 0731 or GPT-5.4: which to pick

Which one for your work

Coding

Intelligence

Agentic work

Reasoning

Math

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