SShogo Bench

GPT-5 Pro

Higher-accuracy GPT-5 tier for tough analysis, coding reviews, and planning

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
44
estimated
Long-horizon autonomy
50
estimated
Knowledge-work agents
33
estimated
Computer use
30
estimated
Security & context learning
43
estimated

Price

Input
$13.64/M
Output
$109.09/M
Cache read
—/M
3:1 blend
$37.50/M
Agent-effective
$18.18/M

Cheapest host for agents: OpenAI at $20.00/M.

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

Agentic benchmarks

BenchmarkScoreNormalisedIndex weightHarness / source
SWE-bench Verifiedest.8%predicted 74
Terminal-Bench 2.0est.8%predicted 58
DeepSWEest.6%predicted 39
FrontierSWEest.5%predicted 10
CursorBenchest.4%predicted 33
GSO-Benchest.4%predicted 19
METR Time Horizonsest.10%predicted 71
Vending-Bench 2est.6%predicted 32
PostTrainBenchest.4%predicted 22
APEX-Agentsest.8%predicted 37
GDPvalest.7%predicted 40
Remote Labor Indexest.5%predicted 3
DeepResearch Benchest.5%predicted 47
OSWorld 2.0est.6%predicted 6
OSWorldest.4%predicted 66
Cybenchest.5%predicted 65
CL-benchest.5%predicted 20

Other benchmarks

BenchmarkCategoryScoreSource
AlgoTuneCoding1.31×source
ARC-AGIReasoning70.2%source
ARC-AGI-2Reasoning18.3%source
Chess PuzzlesReasoning48.0%source
FrontierMath Tiers 1-3Math55.8%source
FrontierMath Tier 4Math19.5%source
Humanity's Last ExamReasoning31.6%source
SimpleBenchReasoning61.6%source

API hosts

Live prices and uptime for every host serving GPT-5 Pro via OpenRouter.

HostInputOutputCache readAgent $/MUptime 24hQuant
OpenAI$15.00$120.00—$20.00100.0%—