SShogo Bench

Claude Sonnet 5.5 vs GPT-5.4

Claude Sonnet 5.5 leads GPT-5.4 by 17.4 points on the Shogo Agent Index. Claude Sonnet 5.5 wins 18 and GPT-5.4 wins 0 of 18 shared benchmarks. Claude Sonnet 5.5 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Sonnet 5.5GPT-5.4
Agentic coding7957
Long-horizon autonomy7561
Knowledge-work agents5740
Computer use6142
Security & context learning6154

Shared benchmarks

BenchmarkClaude Sonnet 5.5GPT-5.4
APEX-Agentsindex75.5%52.4%
Vending-Bench 2index$11,165$6,144
ARC-AGI96.5%93.7%
ARC-AGI-283.9%74.0%
Chess Puzzles59.0%48.0%
CritPt31.4%23.4%
DTBench97.3%94.4%
FrontierMath Tier 480.5%49.0%
FrontierMath Tiers 1-388.8%78.6%
GPQA Diamond95.6%93.3%
Humanity's Last Exam46.1%36.2%
LMCA68.3%52.0%
MirrorCode97.8%23.3%
OTIS Mock AIME100.0%97.8%
ProofBench100.0%56.0%
SciCode61.0%56.6%
SimpleQA Verified46.5%45.1%
WebDev Arena17861463

Specs and price

Claude Sonnet 5.5GPT-5.4
Agent Index69 (#9)51.6 (#33)
Agent $/M$0.84$1.27
Input / output $/M$2.00 / $10.00$2.50 / $15.00
Context1M1.1M
ReleasedSep 28, 2026Mar 5, 2026
WeightsClosedClosed
API hosts54

Don’t pick one. Route.

Shogo can run Claude Sonnet 5.5 and GPT-5.4 side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo