SShogo Bench

Grok 4.6 vs Grok 4.7

Grok 4.6 leads Grok 4.7 by 1.4 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Grok 4.6 wins 9 and Grok 4.7 wins 11 of 20 shared benchmarks. Grok 4.6 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarGrok 4.6Grok 4.7
Agentic coding6158
Long-horizon autonomy6869
Knowledge-work agents4747
Computer use4239
Security & context learning5250

Shared benchmarks

BenchmarkGrok 4.6Grok 4.7
APEX-Agentsindex65.3%70.0%
Vending-Bench 2index$9,047$10,537
FrontierSWEindex25.3%29.5%
CursorBenchindex41.4%46.3%
ARC-AGI87.5%90.2%
ARC-AGI-267.1%61.4%
Chess Puzzles33.0%41.0%
CritPt19.7%18.0%
DTBench97.3%96.0%
FrontierCode48.0%47.6%
FrontierMath Tier 431.7%17.1%
FrontierMath Tiers 1-366.0%53.0%
GPQA Diamond94.0%92.7%
LMCA48.5%49.4%
MirrorCode41.1%54.4%
OTIS Mock AIME99.2%98.1%
ProofBench51.0%26.0%
SciCode56.5%57.8%
SimpleQA Verified49.3%56.0%
WebDev Arena16181638

Specs and price

Grok 4.6Grok 4.7
Agent Index56.2 (#22)54.8 (#27)
Agent $/M$0.98$0.98
Input / output $/M$2.00 / $6.00$2.00 / $6.00
Context500K500K
ReleasedAug 12, 2026Sep 21, 2026
WeightsClosedClosed
API hosts45

Don’t pick one. Route.

Shogo can run Grok 4.6 and Grok 4.7 side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo