SShogo Bench

Claude Sonnet 5 vs GPT-5.5

GPT-5.5 leads Claude Sonnet 5 by 4.3 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Claude Sonnet 5 wins 3 and GPT-5.5 wins 16 of 19 shared benchmarks. Claude Sonnet 5 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Sonnet 5GPT-5.5
Agentic coding5763
Long-horizon autonomy6567
Knowledge-work agents4346
Computer use4043
Security & context learning5155

Shared benchmarks

BenchmarkClaude Sonnet 5GPT-5.5
APEX-Agentsindex54.5%55.1%
DeepSWEindex53.8%67.0%
Vending-Bench 2index$6,378$7,524
GSO-Benchindex37.3%40.2%
Chess Puzzles45.0%63.0%
CritPt16.9%27.1%
DTBench92.5%96.0%
FrontierCode42.7%43.0%
FrontierMath Tier 429.3%72.5%
FrontierMath Tiers 1-365.6%85.3%
GPQA Diamond90.5%94.0%
LMCA50.0%54.3%
MirrorCode63.3%23.3%
OTIS Mock AIME94.7%100.0%
ProofBench77.0%50.0%
SciCode54.3%56.1%
SimpleBench60.6%69.0%
SimpleQA Verified33.7%63.0%
WebDev Arena15371510

Specs and price

Claude Sonnet 5GPT-5.5
Agent Index52.4 (#32)56.7 (#19)
Agent $/M$0.92$2.55
Input / output $/M$2.00 / $10.00$5.00 / $30.00
Context1M1.1M
ReleasedJun 30, 2026Apr 23, 2026
WeightsClosedClosed
API hosts54

Don’t pick one. Route.

Shogo can run Claude Sonnet 5 and GPT-5.5 side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo