SShogo Bench

Claude Opus 4.7 vs Claude Opus 4.8

Claude Opus 4.8 leads Claude Opus 4.7 by 1.7 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Claude Opus 4.7 wins 6 and Claude Opus 4.8 wins 16 of 22 shared benchmarks. Claude Opus 4.7 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Opus 4.7Claude Opus 4.8
Agentic coding6063
Long-horizon autonomy6966
Knowledge-work agents4143
Computer use4447
Security & context learning5255

Shared benchmarks

BenchmarkClaude Opus 4.7Claude Opus 4.8
APEX-Agentsindex49.2%48.9%
OSWorld 2.0index18.2%20.6%
Vending-Bench 2index$10,937$5,787
GSO-Benchindex44.1%47.1%
PostTrainBenchindex24.3%31.0%
ARC-AGI93.5%92.5%
ARC-AGI-275.8%72.1%
Chess Puzzles41.0%52.0%
CritPt12.0%20.9%
DTBench94.7%94.9%
FrontierCode38.5%46.5%
FrontierMath Tier 431.7%56.1%
FrontierMath Tiers 1-370.2%80.0%
GPQA Diamond90.2%91.0%
LMCA52.2%57.5%
MirrorCode43.3%65.6%
OTIS Mock AIME97.8%98.3%
ProofBench54.0%69.0%
SciCode54.5%53.5%
SimpleBench61.7%64.8%
SimpleQA Verified51.7%53.0%
WebDev Arena15671559

Specs and price

Claude Opus 4.7Claude Opus 4.8
Agent Index54.6 (#29)56.3 (#20)
Agent $/M$2.31$2.31
Input / output $/M$5.00 / $25.00$5.00 / $25.00
Context1M1M
ReleasedApr 16, 2026May 28, 2026
WeightsClosedClosed
API hosts56

Don’t pick one. Route.

Shogo can run Claude Opus 4.7 and Claude Opus 4.8 side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo