SShogo Bench

Claude Sonnet 5.5 vs GPT-6 Astra

GPT-6 Astra leads Claude Sonnet 5.5 by 1.9 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Claude Sonnet 5.5 wins 5 and GPT-6 Astra wins 15 of 22 shared benchmarks. Claude Sonnet 5.5 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Sonnet 5.5GPT-6 Astra
Agentic coding7980
Long-horizon autonomy7580
Knowledge-work agents5756
Computer use6167
Security & context learning6163

Shared benchmarks

BenchmarkClaude Sonnet 5.5GPT-6 Astra
APEX-Agentsindex75.5%64.7%
Vending-Bench 2index$11,165$15,515
FrontierSWEindex61.9%65.5%
ARC-AGI96.5%98.5%
ARC-AGI-283.9%95.0%
Chess Puzzles59.0%76.0%
CritPt31.4%31.7%
DTBench97.3%97.3%
FrontierCode52.1%53.3%
FrontierMath Tier 480.5%97.6%
FrontierMath Tiers 1-388.8%93.7%
GPQA Diamond95.6%95.8%
Humanity's Last Exam46.1%54.8%
LMCA68.3%64.4%
MirrorCode97.8%60.0%
OTIS Mock AIME100.0%100.0%
ProofBench100.0%99.0%
SciCode61.0%56.5%
SimpleBench75.9%86.5%
SimpleQA Verified46.5%75.6%
WebDev Arena17861800
WeirdML v319.9%42.2%

Specs and price

Claude Sonnet 5.5GPT-6 Astra
Agent Index69 (#9)70.9 (#4)
Agent $/M$0.84$4.62
Input / output $/M$2.00 / $10.00$10.00 / $50.00
Context1M1.1M
ReleasedSep 28, 2026Sep 4, 2026
WeightsClosedClosed
API hosts56

Don’t pick one. Route.

Shogo can run Claude Sonnet 5.5 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