SShogo Bench

GPT-5.4 Pro

More exact GPT-5.4 tier for demanding professional reasoning and agent tasks

Shogo Agent Index

Unranked

Not ranked yet: 0 agentic benchmarks measured, and we need at least 2 plus an overall capability score. It will be ranked as results are published.

Agentic coding
65
estimated
Long-horizon autonomy
68
estimated
Knowledge-work agents
46
estimated
Computer use
47
estimated
Security & context learning
55
estimated

Price

Input
$27.27/M
Output
$163.64/M
Cache read
—/M
3:1 blend
$61.36/M
Agent-effective
$33.77/M

Cheapest host for agents: Azure at $37.14/M.

Agent-effective = 20 input tokens per output token, 85% cache hits. Why.

Agentic benchmarks

BenchmarkScoreNormalisedIndex weightHarness / source
SWE-bench Verifiedest.8%predicted 83
Terminal-Bench 2.0est.8%predicted 86
DeepSWEest.6%predicted 64
FrontierSWEest.5%predicted 31
CursorBenchest.4%predicted 44
GSO-Benchest.4%predicted 49
METR Time Horizonsest.10%predicted 80
Vending-Bench 2est.6%predicted 72
PostTrainBenchest.4%predicted 30
APEX-Agentsest.8%predicted 58
GDPvalest.7%predicted 57
Remote Labor Indexest.5%predicted 8
DeepResearch Benchest.5%predicted 50
OSWorld 2.0est.6%predicted 19
OSWorldest.4%predicted 87
Cybenchest.5%predicted 85
CL-benchest.5%predicted 25

Other benchmarks

BenchmarkCategoryScoreSource
ARC-AGIReasoning94.5%source
ARC-AGI-2Reasoning83.3%source
Chess PuzzlesReasoning59.0%source
CritPtReasoning30.0%source
FrontierMath Tiers 1-3Math82.5%source
FrontierMath Tier 4Math58.5%source
GPQA DiamondReasoning94.6%source
Humanity's Last ExamReasoning44.3%source
SimpleBenchReasoning74.1%source
SimpleQA VerifiedKnowledge46.3%source

API hosts

Live prices and uptime for every host serving GPT-5.4 Pro via OpenRouter.

HostInputOutputCache readAgent $/MUptime 24hQuant
Azure$30.00$180.00—$37.14100.0%—
OpenAI$30.00$180.00—$37.14100.0%—