Noometry

Model comparison

Grok 4.6 vs Grok 4.7

Grok 4.6 is the stronger model overall, scoring 56.9 to 53.1 on the Noometry Index.

Last verified . 38 shared benchmarks.

Grok 4.6 xAI

56.9

Rank #21 Confirmed

Grok 4.7 xAI

53.1

Rank #37 Confirmed

Summary

  • They share 38 benchmarks with published results for both. Grok 4.6 scores higher in 9 categories and Grok 4.7 in 1 category; 8 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Grok 4.6 leads 61.4 to 49.1.
  • The biggest single-benchmark swing is Furniture Assembly: 40% for Grok 4.6 and 20.8% for Grok 4.7.
  • Both cost about the same: $2 input and $6 output per million tokens.

Side by side

Grok 4.6Grok 4.7
ProviderxAIxAI
Noometry Index56.953.1
Released2026-08-122026-09-21
WeightsProprietaryProprietary
Context window500K500K
Max output500K500K
Input $ / M tokens$2$2
Output $ / M tokens$6$6
Results tracked4939

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

Category by category

Coding Too close to call

Grok 4.6: 58.5 (#16), Grok 4.7: 58.0 (#18)

BenchmarkGrok 4.6Grok 4.7
FrontierCode48%47.6%
CursorBench41.4%46.3%
LMArena WebDev16171639
FrontierSWE25.3%29.5%
SciCode56.5%57.8%
LMArena Coding14651427
DeepSWE67.5%—
WeirdML67.3%—
ALE-Bench1,508—

Agentic & Tool Use Grok 4.6 leads

Grok 4.6: 39.4 (#27), Grok 4.7: 36.7 (#37)

BenchmarkGrok 4.6Grok 4.7
APEX-Agents65.3%54.6%
GDP.pdf17.2%22.8%
Vending-Bench 29,04710,537

Reasoning Grok 4.6 leads

Grok 4.6: 61.4 (#20), Grok 4.7: 49.1 (#40)

BenchmarkGrok 4.6Grok 4.7
NYT Connections (extended)80%76.8%
CritPt19.7%18%
Chess Puzzles40%38%
LMArena Hard Prompts14471413
Mystery Game Puzzles34%29%
DTBench97.3%96%
LMCA48.5%49.4%
Epoch Capabilities Index156.44153.53
ARC-AGI-267.1%—
SimpleBench75.9%—
ARC-AGI-187.5%—
EBR-Bench30.5%—

Math Grok 4.6 leads

Grok 4.6: 67.0 (#24), Grok 4.7: 57.8 (#39)

BenchmarkGrok 4.6Grok 4.7
FrontierMath (Tiers 1-3)66%53%
FrontierMath Tier 431.7%17.1%
OTIS Mock AIME 2024-202599.2%98.1%
ProofBench51%34%
LMArena Math14231407

Knowledge Too close to call

Grok 4.6: 63.3 (#20), Grok 4.7: 62.8 (#22)

BenchmarkGrok 4.6Grok 4.7
GPQA Diamond94%92.7%
SimpleQA Verified49.3%56%
LMArena Expert14671422

Multimodal Grok 4.6 leads

Grok 4.6: 43.6 (#23), Grok 4.7: 35.5 (#87)

BenchmarkGrok 4.6Grok 4.7
LMArena Vision12631228
Blueprint-Bench 233.2%32.5%
Furniture Assembly40%20.8%
LMArena Document1452—

Multilingual Grok 4.6 leads

Grok 4.6: 53.0 (#74), Grok 4.7: 50.8 (#116)

BenchmarkGrok 4.6Grok 4.7
LMArena Non-English14201389
LMArena Chinese14801455
LMArena French14611455
LMArena Russian14221397
LMArena Spanish14041400
LMArena German1431—
LMArena Japanese1376—
LMArena Korean1397—

Instruction Following Grok 4.6 leads

Grok 4.6: 75.4 (#63), Grok 4.7: 74.1 (#105)

BenchmarkGrok 4.6Grok 4.7
LMArena Instruction Following14311404

Long Context Grok 4.6 leads

Grok 4.6: 44.5 (#66), Grok 4.7: 43.1 (#104)

BenchmarkGrok 4.6Grok 4.7
LMArena Longer Query14541413

Writing & Preference Grok 4.7 leads

Grok 4.6: 62.3 (#80), Grok 4.7: 70.0 (#24)

BenchmarkGrok 4.6Grok 4.7
LMArena Text14281399
LMArena Creative Writing14281391
LMArena Multi-Turn14251393
EQ-Bench Creative Writing—2007

Frequently asked questions

Is Grok 4.6 better than Grok 4.7?

Grok 4.6 is the stronger model overall, scoring 56.9 to 53.1 on the Noometry Index.

Which is cheaper, Grok 4.6 or Grok 4.7?

Grok 4.7 is cheaper. It lists at $2 per million input tokens and $6 per million output tokens; Grok 4.6 lists at $2 and $6.

Is Grok 4.6 or Grok 4.7 better for coding?

They score almost the same on coding (58.5 vs 58.0); test both on your own repository before choosing.

Which has the bigger context window?

Both accept 500K tokens.

How many benchmarks do Grok 4.6 and Grok 4.7 share?

38 benchmarks have published results for both models. Grok 4.6 has 49 scored results on Noometry and Grok 4.7 has 39.

Related comparisons

Go deeper