Model comparison

DeepSeek V4 Pro vs GPT-4o

DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 28.6 on the Noometry Index.

Last verified October 10, 2026. 32 shared benchmarks.

DeepSeek V4 Pro

DeepSeek

54.3

Rank #31 Confirmed

GPT-4o OpenAI

28.6

Rank #324 Confirmed

Summary

They share 32 benchmarks with published results for both. DeepSeek V4 Pro scores higher in 9 categories and GPT-4o in 0 categories; 9 gaps are clear of the uncertainty.
The widest gap is in math, where DeepSeek V4 Pro leads 64.8 to 10.6.
The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 98.6% for DeepSeek V4 Pro and 6.4% for GPT-4o.
DeepSeek V4 Pro is cheaper at $0.66 / $1.98 per million input/output tokens, against $2.50 / $10 for GPT-4o.
DeepSeek V4 Pro accepts more context: 1M tokens versus 128K.
DeepSeek V4 Pro has downloadable open weights; the other is API-only.

Side by side

DeepSeek V4 Pro and GPT-4o specifications
	DeepSeek V4 Pro	GPT-4o
Provider	DeepSeek	OpenAI
Noometry Index	54.3	28.6
Released	2026-04-24	2024-05-13
Weights	Open	Proprietary
Context window	1M	128K
Max output	393K	16K
Input $ / M tokens	$0.66	$2.50
Output $ / M tokens	$1.98	$10
Results tracked	48	72

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding DeepSeek V4 Pro leads

DeepSeek V4 Pro: 52.4 (#34), GPT-4o: 24.8 (#328)

Coding benchmarks
Benchmark	DeepSeek V4 Pro	GPT-4o
SWE-bench Verified	77.6%	31%
WeirdML	66.2%	25.1%
LMArena Coding	1470	1297
FrontierCode	28.6%	—
SWE-bench Verified (bash only)	—	21.6%
Aider Polyglot	—	45.3%
LMArena WebDev	1582	—
SciCode	51%	—
GSO	—	0%
BigCodeBench Instruct	—	51.1%
LiveBench Coding	—	51.4%
BigCodeBench Complete	—	61.1%
CadEval	—	26%
ALE-Bench	1,403	—
HumanEval+	—	87.2%
MBPP+	—	72.2%

Agentic & Tool Use DeepSeek V4 Pro leads

DeepSeek V4 Pro: 32.8 (#58), GPT-4o: 21.0 (#141)

Agentic & Tool Use benchmarks
Benchmark	DeepSeek V4 Pro	GPT-4o
APEX-Agents	47.3%	—
GDPval	—	9.9%
TheAgentCompany	—	8.6%
Cybench	—	12.5%
BALROG	—	32.3%
LMArena Search	—	1006
METR Time Horizons	—	40.8%
Vending-Bench 2	3,285	—

Reasoning DeepSeek V4 Pro leads

DeepSeek V4 Pro: 56.5 (#24), GPT-4o: 9.4 (#343)

Reasoning benchmarks
Benchmark	DeepSeek V4 Pro	GPT-4o
ARC-AGI-2	61.3%	0%
ARC-AGI-1	90.5%	4.5%
CritPt	18%	0%
Chess Puzzles	47%	13%
LMArena Hard Prompts	1461	1281
DTBench	93.9%	64.5%
LMCA	45.5%	16.6%
Epoch Capabilities Index	155.31	128.97
ForecastBench	56.1	57.7
SimpleBench	—	17.8%
Kagi LLM Benchmark	53.5%	—
NYT Connections (extended)	91.3%	—
EnigmaEval	—	0.8%
LiveBench Reasoning	—	55.8%
Mystery Game Puzzles	43%	—
LiveBench Data Analysis	—	60.9%
Surface Evolver Bench	40%	—
LiveBench	—	55.3%

Math DeepSeek V4 Pro leads

DeepSeek V4 Pro: 64.8 (#30), GPT-4o: 10.6 (#312)

Math benchmarks
Benchmark	DeepSeek V4 Pro	GPT-4o
FrontierMath (Tiers 1-3)	64.6%	0.4%
OTIS Mock AIME 2024-2025	98.6%	6.4%
LMArena Math	1455	1285
FrontierMath Tier 4	26.8%	—
MathArena Final-Answer Competitions	76.6%	—
ProofBench	50%	—
Omni-MATH	—	29.3%
LiveBench Math	—	49.5%
MATH Level 5	—	53.3%
FrontierMath (Feb 2025 set)	—	0.3%

Knowledge DeepSeek V4 Pro leads

DeepSeek V4 Pro: 59.5 (#31), GPT-4o: 28.8 (#242)

Knowledge benchmarks
Benchmark	DeepSeek V4 Pro	GPT-4o
GPQA Diamond	91.7%	49.2%
SimpleQA Verified	52.9%	26%
Vectara Hallucination Rate	8.6%	9.6%
LMArena Expert	1464	1250
Humanity's Last Exam	—	2.7%
MMLU-Pro	—	71.3%
Confabulations	—	15.3%
GPQA (HELM)	—	52%
MMLU	—	88.1%

Multimodal Not comparable

DeepSeek V4 Pro: —, GPT-4o: 34.5 (#91)

Multimodal benchmarks
Benchmark	DeepSeek V4 Pro	GPT-4o
LMArena Vision	—	1137
Video-MME	—	71.9%
GeoBench	—	71%
VPCT	—	40%
ScienceQA	—	88.5%

Multilingual DeepSeek V4 Pro leads

DeepSeek V4 Pro: 54.4 (#45), GPT-4o: 43.2 (#186)

Multilingual benchmarks
Benchmark	DeepSeek V4 Pro	GPT-4o
LMArena Non-English	1439	1283
LMArena Chinese	1486	1277
LMArena French	1472	1304
LMArena German	1458	1282
LMArena Japanese	1445	1257
LMArena Korean	1447	1234
LMArena Russian	1453	1286
LMArena Spanish	1458	1292

Instruction Following DeepSeek V4 Pro leads

DeepSeek V4 Pro: 76.1 (#47), GPT-4o: 66.6 (#207)

Instruction Following benchmarks
Benchmark	DeepSeek V4 Pro	GPT-4o
LMArena Instruction Following	1448	1278
LiveBench Instruction Following	—	68.6%
IFEval	—	81.7%

Long Context DeepSeek V4 Pro leads

DeepSeek V4 Pro: 45.0 (#51), GPT-4o: 39.4 (#179)

Long Context benchmarks
Benchmark	DeepSeek V4 Pro	GPT-4o
LMArena Longer Query	1458	1289
Fiction.LiveBench	—	66.7%
CL-bench Life	13.5%	—

Writing & Preference DeepSeek V4 Pro leads

DeepSeek V4 Pro: 65.5 (#46), GPT-4o: 52.6 (#166)

Writing & Preference benchmarks
Benchmark	DeepSeek V4 Pro	GPT-4o
LMArena Text	1451	1300
LMArena Creative Writing	1446	1292
LMArena Multi-Turn	1467	1302
Short-Story Creative Writing	—	81.8%
EQ-Bench Creative Writing	1553	—
WildBench	—	82.8%
EQ-Bench 4	1166	—
LiveBench Language	—	47.6%

Frequently asked questions

Is DeepSeek V4 Pro better than GPT-4o?

DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 28.6 on the Noometry Index.

Which is cheaper, DeepSeek V4 Pro or GPT-4o?

DeepSeek V4 Pro is cheaper. It lists at $0.66 per million input tokens and $1.98 per million output tokens; GPT-4o lists at $2.50 and $10.

Is DeepSeek V4 Pro or GPT-4o better for coding?

DeepSeek V4 Pro scores higher on coding benchmarks: 52.4 versus 24.8 in the Noometry coding category.

Which has the bigger context window?

DeepSeek V4 Pro does, with 1M tokens against 128K.

How many benchmarks do DeepSeek V4 Pro and GPT-4o share?

32 benchmarks have published results for both models. DeepSeek V4 Pro has 48 scored results on Noometry and GPT-4o has 72.