SShogo Bench

Claude Opus 4.7 vs GPT-5.4

Claude Opus 4.7 leads GPT-5.4 by 3.0 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Claude Opus 4.7 wins 11 and GPT-5.4 wins 11 of 22 shared benchmarks. GPT-5.4 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Opus 4.7GPT-5.4
Agentic coding6057
Long-horizon autonomy6961
Knowledge-work agents4140
Computer use4442
Security & context learning5254

Shared benchmarks

BenchmarkClaude Opus 4.7GPT-5.4
APEX-Agentsindex49.2%52.4%
SWE-bench Verifiedindex83.5%76.9%
Terminal-Bench 2.0index80.2%81.8%
Vending-Bench 2index$10,937$6,144
GSO-Benchindex44.1%31.4%
PostTrainBenchindex24.3%17.8%
ARC-AGI93.5%93.7%
ARC-AGI-275.8%74.0%
Chess Puzzles41.0%48.0%
CritPt12.0%23.4%
DTBench94.7%94.4%
FrontierMath Tier 431.7%49.0%
FrontierMath Tiers 1-370.2%78.6%
GPQA Diamond90.2%93.3%
Humanity's Last Exam36.2%36.2%
LMCA52.2%52.0%
MirrorCode43.3%23.3%
OTIS Mock AIME97.8%97.8%
ProofBench54.0%56.0%
SciCode54.5%56.6%
SimpleQA Verified51.7%45.1%
WebDev Arena15671463

Specs and price

Claude Opus 4.7GPT-5.4
Agent Index54.6 (#29)51.6 (#33)
Agent $/M$2.31$1.27
Input / output $/M$5.00 / $25.00$2.50 / $15.00
Context1M1.1M
ReleasedApr 16, 2026Mar 5, 2026
WeightsClosedClosed
API hosts54

Don’t pick one. Route.

Shogo can run Claude Opus 4.7 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