SShogo Bench

Claude Sonnet 5 vs GPT-6 Astra

GPT-6 Astra leads Claude Sonnet 5 by 18.5 points on the Shogo Agent Index. Claude Sonnet 5 wins 1 and GPT-6 Astra wins 18 of 19 shared benchmarks. Claude Sonnet 5 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Sonnet 5GPT-6 Astra
Agentic coding5780
Long-horizon autonomy6580
Knowledge-work agents4356
Computer use4067
Security & context learning5163

Shared benchmarks

BenchmarkClaude Sonnet 5GPT-6 Astra
APEX-Agentsindex54.5%64.7%
DeepSWEindex53.8%74.1%
Vending-Bench 2index$6,378$15,515
GSO-Benchindex37.3%79.4%
Chess Puzzles45.0%76.0%
CritPt16.9%31.7%
DTBench92.5%97.3%
FrontierCode42.7%53.3%
FrontierMath Tier 429.3%97.6%
FrontierMath Tiers 1-365.6%93.7%
GPQA Diamond90.5%95.8%
LMCA50.0%64.4%
MirrorCode63.3%60.0%
OTIS Mock AIME94.7%100.0%
ProofBench77.0%99.0%
SciCode54.3%56.5%
SimpleBench60.6%86.5%
SimpleQA Verified33.7%75.6%
WebDev Arena15371800

Specs and price

Claude Sonnet 5GPT-6 Astra
Agent Index52.4 (#32)70.9 (#4)
Agent $/M$0.92$4.62
Input / output $/M$2.00 / $10.00$10.00 / $50.00
Context1M1.1M
ReleasedJun 30, 2026Sep 4, 2026
WeightsClosedClosed
API hosts56

Don’t pick one. Route.

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

Try both in Shogo