SShogo Bench

Claude Opus 4.8 vs Grok 4.6

Claude Opus 4.8 leads Grok 4.6 by 0.1 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Claude Opus 4.8 wins 10 and Grok 4.6 wins 10 of 20 shared benchmarks. Grok 4.6 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Opus 4.8Grok 4.6
Agentic coding6361
Long-horizon autonomy6668
Knowledge-work agents4347
Computer use4742
Security & context learning5552

Shared benchmarks

BenchmarkClaude Opus 4.8Grok 4.6
APEX-Agentsindex48.9%65.3%
DeepSWEindex59.0%67.5%
Vending-Bench 2index$5,787$9,047
ARC-AGI92.5%87.5%
ARC-AGI-272.1%67.1%
Chess Puzzles52.0%33.0%
CritPt20.9%19.7%
DTBench94.9%97.3%
FrontierCode46.5%48.0%
FrontierMath Tier 456.1%31.7%
FrontierMath Tiers 1-380.0%66.0%
GPQA Diamond91.0%94.0%
LMCA57.5%48.5%
MirrorCode65.6%41.1%
OTIS Mock AIME98.3%99.2%
ProofBench69.0%51.0%
SciCode53.5%56.5%
SimpleBench64.8%75.9%
SimpleQA Verified53.0%49.3%
WebDev Arena15591618

Specs and price

Claude Opus 4.8Grok 4.6
Agent Index56.3 (#20)56.2 (#22)
Agent $/M$2.31$0.98
Input / output $/M$5.00 / $25.00$2.00 / $6.00
Context1M500K
ReleasedMay 28, 2026Aug 12, 2026
WeightsClosedClosed
API hosts64

Don’t pick one. Route.

Shogo can run Claude Opus 4.8 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