SShogo Bench

Claude Opus 4.7 vs Grok 4.6

Grok 4.6 leads Claude Opus 4.7 by 1.6 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Claude Opus 4.7 wins 9 and Grok 4.6 wins 9 of 19 shared benchmarks. Grok 4.6 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Opus 4.7Grok 4.6
Agentic coding6061
Long-horizon autonomy6968
Knowledge-work agents4147
Computer use4442
Security & context learning5252

Shared benchmarks

BenchmarkClaude Opus 4.7Grok 4.6
APEX-Agentsindex49.2%65.3%
Vending-Bench 2index$10,937$9,047
ARC-AGI93.5%87.5%
ARC-AGI-275.8%67.1%
Chess Puzzles41.0%33.0%
CritPt12.0%19.7%
DTBench94.7%97.3%
FrontierCode38.5%48.0%
FrontierMath Tier 431.7%31.7%
FrontierMath Tiers 1-370.2%66.0%
GPQA Diamond90.2%94.0%
LMCA52.2%48.5%
MirrorCode43.3%41.1%
OTIS Mock AIME97.8%99.2%
ProofBench54.0%51.0%
SciCode54.5%56.5%
SimpleBench61.7%75.9%
SimpleQA Verified51.7%49.3%
WebDev Arena15671618

Specs and price

Claude Opus 4.7Grok 4.6
Agent Index54.6 (#29)56.2 (#22)
Agent $/M$2.31$0.98
Input / output $/M$5.00 / $25.00$2.00 / $6.00
Context1M500K
ReleasedApr 16, 2026Aug 12, 2026
WeightsClosedClosed
API hosts54

Don’t pick one. Route.

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

Try both in Shogo