SShogo Bench

Claude Sonnet 5.5 vs GPT-5.6 Sol

Claude Sonnet 5.5 leads GPT-5.6 Sol by 7.2 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Claude Sonnet 5.5 wins 14 and GPT-5.6 Sol wins 7 of 22 shared benchmarks. Claude Sonnet 5.5 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Sonnet 5.5GPT-5.6 Sol
Agentic coding7971
Long-horizon autonomy7572
Knowledge-work agents5746
Computer use6153
Security & context learning6158

Shared benchmarks

BenchmarkClaude Sonnet 5.5GPT-5.6 Sol
APEX-Agentsindex75.5%51.4%
Vending-Bench 2index$11,165$9,619
FrontierSWEindex61.9%32.2%
CursorBenchindex55.5%41.7%
ARC-AGI96.5%97.5%
ARC-AGI-283.9%92.5%
Chess Puzzles59.0%68.0%
CritPt31.4%32.3%
DTBench97.3%96.0%
FrontierCode52.1%47.5%
FrontierMath Tier 480.5%82.9%
FrontierMath Tiers 1-388.8%89.1%
GPQA Diamond95.6%93.5%
LMCA68.3%59.2%
MirrorCode97.8%20.0%
OTIS Mock AIME100.0%100.0%
ProofBench100.0%83.0%
SciCode61.0%57.8%
SimpleBench75.9%71.7%
SimpleQA Verified46.5%69.7%
WebDev Arena17861617
WeirdML v319.9%15.0%

Specs and price

Claude Sonnet 5.5GPT-5.6 Sol
Agent Index69 (#9)61.8 (#14)
Agent $/M$0.84$1.85
Input / output $/M$2.00 / $10.00$4.00 / $20.00
Context1M1.1M
ReleasedSep 28, 2026Aug 7, 2026
WeightsClosedClosed
API hosts55

Don’t pick one. Route.

Shogo can run Claude Sonnet 5.5 and GPT-5.6 Sol side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo