SShogo Bench

Grok 4

Grok 4 from xAI.

Shogo Agent Index

Supported#68 of 92
30.9range 22.4–39.4 · 35% of weight measured
Agentic coding
32
23% measured
Long-horizon autonomy
42
50% measured
Knowledge-work agents
25
48% measured
Computer use
22
estimated
Security & context learning
30
50% measured

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 68
Terminal-Bench 2.027.2%8% OpenHands
DeepSWEest.6%predicted 28
FrontierSWEest.5%predicted 5
CursorBenchest.4%predicted 28
GSO-Benchest.4%predicted 11
METR Time Horizons66.6%10%source
Vending-Bench 2est.6%predicted 17
PostTrainBenchest.4%predicted 18
APEX-Agentsest.8%predicted 28
GDPval21.1%7%source
Remote Labor Indexest.5%predicted 1
DeepResearch Bench47.3%5%True
OSWorld 2.0est.6%predicted 3
OSWorldest.4%predicted 51
Cybench43.0%5%source
CL-benchest.5%predicted 18

Other benchmarks

BenchmarkCategoryScoreSource
Aider PolyglotCoding79.6%source
ARC-AGIReasoning66.7%source
ARC-AGI-2Reasoning16.0%source
BALROGLong-horizon43.6%source
Fiction.LiveBenchLong context94.4%source
GeoBenchMultimodal45.0%source
GPQA DiamondReasoning87.0%source
OTIS Mock AIMEMath84.0%source
SimpleBenchReasoning60.5%source