SShogo Bench

Claude Opus 4.8 vs Claude Opus 5

Claude Opus 5 leads Claude Opus 4.8 by 9.1 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Claude Opus 4.8 wins 1 and Claude Opus 5 wins 21 of 22 shared benchmarks. Claude Opus 4.8 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Opus 4.8Claude Opus 5
Agentic coding6374
Long-horizon autonomy6674
Knowledge-work agents4352
Computer use4756
Security & context learning5559

Shared benchmarks

BenchmarkClaude Opus 4.8Claude Opus 5
APEX-Agentsindex48.9%65.8%
DeepSWEindex59.0%73.6%
OSWorld 2.0index20.6%31.4%
Vending-Bench 2index$5,787$11,182
PostTrainBenchindex31.0%35.7%
ARC-AGI92.5%97.5%
ARC-AGI-272.1%90.4%
Chess Puzzles52.0%43.0%
CritPt20.9%29.1%
DTBench94.9%97.9%
FrontierCode46.5%53.4%
FrontierMath Tier 456.1%73.2%
FrontierMath Tiers 1-380.0%85.6%
GPQA Diamond91.0%93.9%
LMCA57.5%64.5%
MirrorCode65.6%87.8%
OTIS Mock AIME98.3%98.9%
ProofBench69.0%99.0%
SciCode53.5%56.4%
SimpleBench64.8%80.6%
SimpleQA Verified53.0%59.9%
WebDev Arena15591687

Specs and price

Claude Opus 4.8Claude Opus 5
Agent Index56.3 (#20)65.4 (#11)
Agent $/M$2.31$2.31
Input / output $/M$5.00 / $25.00$5.00 / $25.00
Context1M1M
ReleasedMay 28, 2026Jul 24, 2026
WeightsClosedClosed
API hosts66

Don’t pick one. Route.

Shogo can run Claude Opus 4.8 and Claude Opus 5 side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo