Noometry

Model comparison

GPT-6 Astra vs Grok 4 Fast

GPT-6 Astra is the stronger model overall, scoring 70.3 to 39.0 on the Noometry Index.

Last verified . 21 shared benchmarks.

GPT-6 Astra OpenAI

70.3

Rank #1 Confirmed

Grok 4 Fast xAI

39.0

Rank #164 Confirmed

Summary

  • They share 21 benchmarks with published results for both. GPT-6 Astra scores higher in 8 categories and Grok 4 Fast in 1 category; 9 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where GPT-6 Astra leads 85.1 to 22.5.
  • The biggest single-benchmark swing is ARC-AGI-2: 95% for GPT-6 Astra and 5.3% for Grok 4 Fast.

Side by side

GPT-6 AstraGrok 4 Fast
ProviderOpenAIxAI
Noometry Index70.339.0
Released2026-09-032025-09-19
WeightsProprietaryProprietary
Context window1.05M—
Max output128K—
Input $ / M tokens$10—
Output $ / M tokens$50—
Results tracked5130

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

Category by category

Coding GPT-6 Astra leads

GPT-6 Astra: 70.5 (#3), Grok 4 Fast: 32.3 (#264)

BenchmarkGPT-6 AstraGrok 4 Fast
LMArena WebDev17921160
WeirdML93.6%42.9%
LMArena Coding14881429
DeepSWE74.1%—
FrontierCode53.3%—
FrontierSWE65.5%—
SciCode56.5%—
MirrorCode46.7%—
ALE-Bench2,951—

Agentic & Tool Use GPT-6 Astra leads

GPT-6 Astra: 52.3 (#3), Grok 4 Fast: 29.3 (#80)

BenchmarkGPT-6 AstraGrok 4 Fast
APEX-Agents64.7%—
Remote Labor Index20.8%—
τ²-bench Banking—15.7%
Cybench—30%
BALROG68.3%—
LMArena Search—1171
Vending-Bench 215,515—

Reasoning GPT-6 Astra leads

GPT-6 Astra: 85.1 (#1), Grok 4 Fast: 22.5 (#179)

BenchmarkGPT-6 AstraGrok 4 Fast
ARC-AGI-295%5.3%
ARC-AGI-198.5%48.5%
LMArena Hard Prompts14641413
DTBench97.3%82.7%
Epoch Capabilities Index166.6144.21
Kagi LLM Benchmark—66.1%
NYT Connections (extended)98.1%—
CritPt31.7%—
Chess Puzzles72%—
EBR-Bench76.2%—
Mystery Game Puzzles84%—
LMCA64.4%—
Bench to the Future 30.14—
ForecastBench—60.5

Math GPT-6 Astra leads

GPT-6 Astra: 93.4 (#1), Grok 4 Fast: 36.1 (#126)

BenchmarkGPT-6 AstraGrok 4 Fast
LMArena Math14631419
FrontierMath (Tiers 1-3)93.7%—
FrontierMath Tier 497.6%—
OTIS Mock AIME 2024-2025100%—
ProofBench99%—
FrontierMath Erdős2.9%—

Knowledge GPT-6 Astra leads

GPT-6 Astra: 75.3 (#1), Grok 4 Fast: 31.5 (#211)

BenchmarkGPT-6 AstraGrok 4 Fast
Vectara Hallucination Rate8.7%19.7%
LMArena Expert14861411
GPQA Diamond95.8%—
Humanity's Last Exam54.8%—
SimpleQA Verified75.6%—

Multimodal Not comparable

GPT-6 Astra: 55.1 (#3), Grok 4 Fast: —

BenchmarkGPT-6 AstraGrok 4 Fast
LMArena Vision1280—
Blueprint-Bench 249.7%—
Furniture Assembly80%—
LMArena Document1468—

Multilingual GPT-6 Astra leads

GPT-6 Astra: 54.0 (#53), Grok 4 Fast: 51.5 (#106)

BenchmarkGPT-6 AstraGrok 4 Fast
LMArena Non-English14311397
LMArena Chinese14951458
LMArena French14671433
LMArena Russian14341387
LMArena Spanish14121414
LMArena German—1383
LMArena Japanese—1358
LMArena Korean—1359

Instruction Following GPT-6 Astra leads

GPT-6 Astra: 76.6 (#37), Grok 4 Fast: 73.4 (#116)

BenchmarkGPT-6 AstraGrok 4 Fast
LMArena Instruction Following14541388

Long Context Grok 4 Fast leads

GPT-6 Astra: 44.7 (#57), Grok 4 Fast: 63.2 (#3)

BenchmarkGPT-6 AstraGrok 4 Fast
LMArena Longer Query14601415
Fiction.LiveBench—94.4%

Writing & Preference GPT-6 Astra leads

GPT-6 Astra: 75.7 (#7), Grok 4 Fast: 60.2 (#96)

BenchmarkGPT-6 AstraGrok 4 Fast
LMArena Text14431407
LMArena Creative Writing14231388
LMArena Multi-Turn14571414
EQ-Bench Creative Writing2173—

Frequently asked questions

Is GPT-6 Astra better than Grok 4 Fast?

GPT-6 Astra is the stronger model overall, scoring 70.3 to 39.0 on the Noometry Index.

Is GPT-6 Astra or Grok 4 Fast better for coding?

GPT-6 Astra scores higher on coding benchmarks: 70.5 versus 32.3 in the Noometry coding category.

How many benchmarks do GPT-6 Astra and Grok 4 Fast share?

21 benchmarks have published results for both models. GPT-6 Astra has 51 scored results on Noometry and Grok 4 Fast has 30.

Related comparisons

Go deeper