Noometry

Model comparison

GPT-5 Nano vs Grok-3 mini

Grok-3 mini is the stronger model overall, scoring 41.0 to 33.4 on the Noometry Index.

Last verified . 30 shared benchmarks.

GPT-5 Nano OpenAI

33.4

Rank #234 Confirmed

Grok-3 mini xAI

41.0

Rank #130 Confirmed

Summary

  • They share 30 benchmarks with published results for both. GPT-5 Nano scores higher in 1 category and Grok-3 mini in 7 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where Grok-3 mini leads 52.7 to 39.4.
  • The biggest single-benchmark swing is Omni-MATH: 54.6% for GPT-5 Nano and 31.8% for Grok-3 mini.

Side by side

GPT-5 NanoGrok-3 mini
ProviderOpenAIxAI
Noometry Index33.441.0
Released2025-08-072025-04-09
WeightsProprietaryProprietary
Context window400K—
Max output128K—
Input $ / M tokens$0.05—
Output $ / M tokens$0.40—
Results tracked4835

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

Category by category

Coding Grok-3 mini leads

GPT-5 Nano: 33.3 (#248), Grok-3 mini: 40.4 (#126)

BenchmarkGPT-5 NanoGrok-3 mini
WeirdML38.1%42.6%
LMArena Coding13511380
SWE-bench Verified (bash only)34.8%—
Aider Polyglot—49.3%
ALE-Bench718.67—

Agentic & Tool Use Not comparable

GPT-5 Nano: 25.6 (#99), Grok-3 mini: —

BenchmarkGPT-5 NanoGrok-3 mini
Terminal-Bench21.8%—
Berkeley Function Calling Leaderboard51.5%—

Reasoning GPT-5 Nano leads

GPT-5 Nano: 16.4 (#291), Grok-3 mini: 13.8 (#323)

BenchmarkGPT-5 NanoGrok-3 mini
ARC-AGI-22.6%0.4%
Kagi LLM Benchmark62.2%61.3%
ARC-AGI-120.7%16.5%
LMArena Hard Prompts13291374
Epoch Capabilities Index139.39140.35
Chess Puzzles27%—
Mystery Game Puzzles9%—
DTBench62.7%—
LMCA7.9%—
ForecastBench59.1—

Math Grok-3 mini leads

GPT-5 Nano: 29.1 (#221), Grok-3 mini: 40.3 (#78)

BenchmarkGPT-5 NanoGrok-3 mini
OTIS Mock AIME 2024-202581.1%77.8%
Omni-MATH54.6%31.8%
LMArena Math13181386
MATH Level 595.2%90.9%
FrontierMath (Feb 2025 set)8.3%5.9%
FrontierMath (Tiers 1-3)20%—
FrontierMath Tier 42.4%—
ProofBench12%—
FrontierMath Tier 4 (v1)2.1%—

Knowledge Grok-3 mini leads

GPT-5 Nano: 35.8 (#166), Grok-3 mini: 46.3 (#75)

BenchmarkGPT-5 NanoGrok-3 mini
GPQA Diamond69.4%76.3%
MMLU-Pro77.8%79.9%
GPQA (HELM)67.9%67.5%
LMArena Expert13211395
SimpleQA Verified11.7%—
Confabulations—10.8%
Vectara Hallucination Rate10.5%—

Multimodal Not comparable

GPT-5 Nano: 30.7 (#102), Grok-3 mini: —

BenchmarkGPT-5 NanoGrok-3 mini
LMArena Vision1159—
VPCT37.2%—

Multilingual Grok-3 mini leads

GPT-5 Nano: 45.5 (#167), Grok-3 mini: 48.3 (#141)

BenchmarkGPT-5 NanoGrok-3 mini
LMArena Non-English13131352
LMArena Chinese13551387
LMArena German13291350
LMArena Korean12691336
LMArena Russian12961352
LMArena Spanish13601379
LMArena French—1357
LMArena Japanese—1343

Instruction Following Grok-3 mini leads

GPT-5 Nano: 75.2 (#72), Grok-3 mini: 78.7 (#8)

BenchmarkGPT-5 NanoGrok-3 mini
IFEval93.2%95.1%
LMArena Instruction Following13061356

Long Context Grok-3 mini leads

GPT-5 Nano: 31.3 (#275), Grok-3 mini: 41.0 (#142)

BenchmarkGPT-5 NanoGrok-3 mini
Fiction.LiveBench44.4%66.7%
LMArena Longer Query13131372

Writing & Preference Grok-3 mini leads

GPT-5 Nano: 39.4 (#244), Grok-3 mini: 52.7 (#163)

BenchmarkGPT-5 NanoGrok-3 mini
LMArena Text13201370
LMArena Creative Writing12501342
WildBench80.6%65.1%
LMArena Multi-Turn13111355
Short-Story Creative Writing—73.5%
EQ-Bench Creative Writing705—

Frequently asked questions

Is GPT-5 Nano better than Grok-3 mini?

Grok-3 mini is the stronger model overall, scoring 41.0 to 33.4 on the Noometry Index.

Is GPT-5 Nano or Grok-3 mini better for coding?

Grok-3 mini scores higher on coding benchmarks: 40.4 versus 33.3 in the Noometry coding category.

How many benchmarks do GPT-5 Nano and Grok-3 mini share?

30 benchmarks have published results for both models. GPT-5 Nano has 48 scored results on Noometry and Grok-3 mini has 35.

Related comparisons

Go deeper