SShogo Bench

GPT-5 Nano

Tiny GPT-5 lane for routing, extraction, classification, and bulk jobs

Shogo Agent Index

Unranked

Not ranked yet: 1 agentic benchmark measured, and we need at least 2 plus an overall capability score. It will be ranked as results are published.

Agentic coding
23
23% measured
Long-horizon autonomy
32
estimated
Knowledge-work agents
20
estimated
Computer use
11
estimated
Security & context learning
22
estimated

Price

Input
$0.045/M
Output
$0.36/M
Cache read
$0.004/M
3:1 blend
$0.13/M
Agent-effective
$0.030/M

Cheapest host for agents: OpenAI at $0.020/M.

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

Agentic benchmarks

BenchmarkScoreNormalisedIndex weightHarness / source
SWE-bench Verifiedest.8%predicted 56
Terminal-Bench 2.021.8%8%spoox-o-m
DeepSWEest.6%predicted 14
FrontierSWEest.5%predicted 2
CursorBenchest.4%predicted 21
GSO-Benchest.4%predicted 3
METR Time Horizonsest.10%predicted 56
Vending-Bench 2est.6%predicted 4
PostTrainBenchest.4%predicted 13
APEX-Agentsest.8%predicted 15
GDPvalest.7%predicted 21
Remote Labor Indexest.5%predicted 1
DeepResearch Benchest.5%predicted 43
OSWorld 2.0est.6%predicted 1
OSWorldest.4%predicted 25
Cybenchest.5%predicted 29
CL-benchest.5%predicted 14

Other benchmarks

BenchmarkCategoryScoreSource
ARC-AGIReasoning20.7%source
ARC-AGI-2Reasoning2.6%source
Chess PuzzlesReasoning22.0%source
DTBenchKnowledge62.7%source
Fiction.LiveBenchLong context44.4%source
FrontierMath Tiers 1-3Math20.0%source
FrontierMath Tier 4Math2.4%source
GPQA DiamondReasoning69.4%source
LMCAKnowledge7.9%source
MATH Level 5Math95.2%source
OTIS Mock AIMEMath81.1%source
ProofBenchMath12.0%source
SimpleQA VerifiedKnowledge11.7%source
VPCTMultimodal37.2%source

API hosts

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

HostInputOutputCache readAgent $/MUptime 24hQuant
OpenAI$0.025$0.20$0.003$0.020100.0%—
Azure$0.050$0.40$0.010$0.030100.0%—
OpenAI$0.050$0.40$0.005$0.030100.0%—