SShogo Bench

GPT-5.4 vs GPT-5.5

GPT-5.5 leads GPT-5.4 by 5.1 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. GPT-5.4 wins 2 and GPT-5.5 wins 20 of 23 shared benchmarks. GPT-5.4 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarGPT-5.4GPT-5.5
Agentic coding5763
Long-horizon autonomy6167
Knowledge-work agents4046
Computer use4243
Security & context learning5455

Shared benchmarks

BenchmarkGPT-5.4GPT-5.5
APEX-Agentsindex52.4%55.1%
SWE-bench Verifiedindex76.9%80.6%
Terminal-Bench 2.0index81.8%84.7%
DeepSWEindex51.8%67.0%
Vending-Bench 2index$6,144$7,524
DeepResearch Benchindex35.1%54.0%
GSO-Benchindex31.4%40.2%
PostTrainBenchindex17.8%23.9%
ARC-AGI93.7%95.0%
ARC-AGI-274.0%85.0%
Chess Puzzles48.0%63.0%
CritPt23.4%27.1%
DTBench94.4%96.0%
FrontierMath Tier 449.0%72.5%
FrontierMath Tiers 1-378.6%85.3%
GPQA Diamond93.3%94.0%
LMCA52.0%54.3%
MirrorCode23.3%23.3%
OTIS Mock AIME97.8%100.0%
ProofBench56.0%50.0%
SciCode56.6%56.1%
SimpleQA Verified45.1%63.0%
WebDev Arena14631510

Specs and price

GPT-5.4GPT-5.5
Agent Index51.6 (#33)56.7 (#19)
Agent $/M$1.27$2.55
Input / output $/M$2.50 / $15.00$5.00 / $30.00
Context1.1M1.1M
ReleasedMar 5, 2026Apr 23, 2026
WeightsClosedClosed
API hosts44

Don’t pick one. Route.

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

Try both in Shogo