SShogo Bench

Claude Mythos Preview

Claude Mythos Preview from Anthropic.

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
86
estimated
Long-horizon autonomy
81
50% measured
Knowledge-work agents
65
estimated
Computer use
76
estimated
Security & context learning
65
estimated

Price

No public API price.

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

Agentic benchmarks

BenchmarkScoreNormalisedIndex weightHarness / source
SWE-bench Verifiedest.8%predicted 92
Terminal-Bench 2.0est.8%predicted 98
DeepSWEest.6%predicted 88
FrontierSWEest.5%predicted 76
CursorBenchest.4%predicted 61
GSO-Benchest.4%predicted 87
METR Time Horizons85.2%10%source
Vending-Bench 2est.6%predicted 97
PostTrainBenchest.4%predicted 45
APEX-Agentsest.8%predicted 82
GDPvalest.7%predicted 77
Remote Labor Indexest.5%predicted 32
DeepResearch Benchest.5%predicted 54
OSWorld 2.0est.6%predicted 61
OSWorldest.4%predicted 98
Cybenchest.5%predicted 96
CL-benchest.5%predicted 34

Other benchmarks

BenchmarkCategoryScoreSource
ExploitBenchSecurity & context73.8%source