SShogo Bench

Claude Sonnet 5 vs GPT-6 Sol

GPT-6 Sol leads Claude Sonnet 5 by 11.2 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Claude Sonnet 5 wins 2 and GPT-6 Sol wins 15 of 17 shared benchmarks. Claude Sonnet 5 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Sonnet 5GPT-6 Sol
Agentic coding5772
Long-horizon autonomy6576
Knowledge-work agents4348
Computer use4054
Security & context learning5159

Shared benchmarks

BenchmarkClaude Sonnet 5GPT-6 Sol
APEX-Agentsindex54.5%54.3%
Vending-Bench 2index$6,378$14,428
Chess Puzzles45.0%56.0%
CritPt16.9%30.9%
DTBench92.5%97.3%
FrontierCode42.7%49.3%
FrontierMath Tier 429.3%90.0%
FrontierMath Tiers 1-365.6%89.8%
GPQA Diamond90.5%94.3%
LMCA50.0%59.1%
MirrorCode63.3%48.9%
OTIS Mock AIME94.7%100.0%
ProofBench77.0%83.0%
SciCode54.3%57.6%
SimpleBench60.6%73.1%
SimpleQA Verified33.7%60.7%
WebDev Arena15371689

Specs and price

Claude Sonnet 5GPT-6 Sol
Agent Index52.4 (#32)63.6 (#13)
Agent $/M$0.92$0.92
Input / output $/M$2.00 / $10.00$2.00 / $10.00
Context1M1.1M
ReleasedJun 30, 2026Sep 22, 2026
WeightsClosedClosed
API hosts55

Don’t pick one. Route.

Shogo can run Claude Sonnet 5 and GPT-6 Sol side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo