SShogo Bench

GPT-5.2 Pro

Higher-accuracy GPT-5.2 variant for tougher reasoning and review workflows

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
58
estimated
Long-horizon autonomy
62
estimated
Knowledge-work agents
42
estimated
Computer use
41
estimated
Security & context learning
52
estimated

Price

Input
$19.09/M
Output
$152.73/M
Cache read
—/M
3:1 blend
$52.50/M
Agent-effective
$25.45/M

Cheapest host for agents: OpenAI at $28.00/M.

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

Agentic benchmarks

BenchmarkScoreNormalisedIndex weightHarness / source
SWE-bench Verifiedest.8%predicted 80
Terminal-Bench 2.0est.8%predicted 79
DeepSWEest.6%predicted 56
FrontierSWEest.5%predicted 22
CursorBenchest.4%predicted 41
GSO-Benchest.4%predicted 37
METR Time Horizonsest.10%predicted 78
Vending-Bench 2est.6%predicted 60
PostTrainBenchest.4%predicted 27
APEX-Agentsest.8%predicted 51
GDPvalest.7%predicted 52
Remote Labor Indexest.5%predicted 5
DeepResearch Benchest.5%predicted 49
OSWorld 2.0est.6%predicted 13
OSWorldest.4%predicted 82
Cybenchest.5%predicted 80
CL-benchest.5%predicted 23

Other benchmarks

BenchmarkCategoryScoreSource
ARC-AGIReasoning90.5%source
ARC-AGI-2Reasoning54.2%source
Chess PuzzlesReasoning60.0%source
FrontierMath Tiers 1-3Math74.0%source
FrontierMath Tier 4Math46.0%source
SimpleBenchReasoning57.4%source

API hosts

Live prices and uptime for every host serving GPT-5.2 Pro via OpenRouter.

HostInputOutputCache readAgent $/MUptime 24hQuant
OpenAI$21.00$168.00—$28.00100.0%—