Noometry

Model comparison

DeepSeek V4 Pro vs Grok 4.3

DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 43.8 on the Noometry Index.

Last verified . 36 shared benchmarks.

DeepSeek V4 Pro DeepSeek

54.3

Rank #31 Confirmed

Grok 4.3 xAI

43.8

Rank #86 Confirmed

Summary

  • They share 36 benchmarks with published results for both. DeepSeek V4 Pro scores higher in 9 categories and Grok 4.3 in 0 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where DeepSeek V4 Pro leads 56.5 to 35.9.
  • The biggest single-benchmark swing is ProofBench: 50% for DeepSeek V4 Pro and 11% for Grok 4.3.
  • DeepSeek V4 Pro is cheaper at $0.66 / $1.98 per million input/output tokens, against $1.25 / $2.50 for Grok 4.3.
  • DeepSeek V4 Pro has downloadable open weights; the other is API-only.

Side by side

DeepSeek V4 ProGrok 4.3
ProviderDeepSeekxAI
Noometry Index54.343.8
Released2026-04-242026-04-17
WeightsOpenProprietary
Context window1M1M
Max output393K30K
Input $ / M tokens$0.66$1.25
Output $ / M tokens$1.98$2.50
Results tracked4840

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), Grok 4.3: 41.6 (#121)

BenchmarkDeepSeek V4 ProGrok 4.3
LMArena WebDev15821357
SciCode51%47.3%
WeirdML66.2%49.9%
LMArena Coding14701415
ALE-Bench1,403944.17
SWE-bench Verified77.6%—
FrontierCode28.6%—

Agentic & Tool Use DeepSeek V4 Pro leads

DeepSeek V4 Pro: 32.8 (#58), Grok 4.3: 27.7 (#99)

BenchmarkDeepSeek V4 ProGrok 4.3
Vending-Bench 23,28535.26
APEX-Agents47.3%—
GDP.pdf—8%
LMArena Search—1165

Reasoning DeepSeek V4 Pro leads

DeepSeek V4 Pro: 56.5 (#24), Grok 4.3: 35.9 (#68)

BenchmarkDeepSeek V4 ProGrok 4.3
NYT Connections (extended)91.3%55.2%
CritPt18%8%
Chess Puzzles47%25%
LMArena Hard Prompts14611396
DTBench93.9%90.7%
LMCA45.5%38.3%
Epoch Capabilities Index155.31149.16
ForecastBench56.160.3
ARC-AGI-261.3%—
Kagi LLM Benchmark53.5%—
ARC-AGI-190.5%—
Mystery Game Puzzles43%—
Surface Evolver Bench40%—

Math DeepSeek V4 Pro leads

DeepSeek V4 Pro: 64.8 (#30), Grok 4.3: 46.0 (#74)

BenchmarkDeepSeek V4 ProGrok 4.3
FrontierMath (Tiers 1-3)64.6%42.8%
FrontierMath Tier 426.8%14.6%
OTIS Mock AIME 2024-202598.6%93.3%
ProofBench50%11%
LMArena Math14551388
MathArena Final-Answer Competitions76.6%—

Knowledge DeepSeek V4 Pro leads

DeepSeek V4 Pro: 59.5 (#31), Grok 4.3: 52.5 (#62)

BenchmarkDeepSeek V4 ProGrok 4.3
GPQA Diamond91.7%88.8%
SimpleQA Verified52.9%33.2%
LMArena Expert14641385
Vectara Hallucination Rate8.6%—

Multimodal Not comparable

DeepSeek V4 Pro: —, Grok 4.3: 31.6 (#104)

BenchmarkDeepSeek V4 ProGrok 4.3
LMArena Vision—1229
Blueprint-Bench 2—0%

Multilingual DeepSeek V4 Pro leads

DeepSeek V4 Pro: 54.4 (#45), Grok 4.3: 50.5 (#120)

BenchmarkDeepSeek V4 ProGrok 4.3
LMArena Non-English14391385
LMArena Chinese14861422
LMArena French14721412
LMArena German14581395
LMArena Japanese14451379
LMArena Korean14471356
LMArena Russian14531399
LMArena Spanish14581398

Instruction Following DeepSeek V4 Pro leads

DeepSeek V4 Pro: 76.1 (#47), Grok 4.3: 72.1 (#140)

BenchmarkDeepSeek V4 ProGrok 4.3
LMArena Instruction Following14481366

Long Context DeepSeek V4 Pro leads

DeepSeek V4 Pro: 45.0 (#51), Grok 4.3: 42.5 (#123)

BenchmarkDeepSeek V4 ProGrok 4.3
LMArena Longer Query14581393
CL-bench Life13.5%—

Writing & Preference DeepSeek V4 Pro leads

DeepSeek V4 Pro: 65.5 (#46), Grok 4.3: 58.5 (#118)

BenchmarkDeepSeek V4 ProGrok 4.3
LMArena Text14511397
LMArena Creative Writing14461380
EQ-Bench 411661075
LMArena Multi-Turn14671406
EQ-Bench Creative Writing1553—

Frequently asked questions

Is DeepSeek V4 Pro better than Grok 4.3?

DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 43.8 on the Noometry Index.

Which is cheaper, DeepSeek V4 Pro or Grok 4.3?

DeepSeek V4 Pro is cheaper. It lists at $0.66 per million input tokens and $1.98 per million output tokens; Grok 4.3 lists at $1.25 and $2.50.

Is DeepSeek V4 Pro or Grok 4.3 better for coding?

DeepSeek V4 Pro scores higher on coding benchmarks: 52.4 versus 41.6 in the Noometry coding category.

Which has the bigger context window?

Both accept 1M tokens.

How many benchmarks do DeepSeek V4 Pro and Grok 4.3 share?

36 benchmarks have published results for both models. DeepSeek V4 Pro has 48 scored results on Noometry and Grok 4.3 has 40.

Related comparisons

Go deeper