SShogo Bench

Grok Build 0.1

Fast Grok coding model tuned for agentic engineering and iterative edits

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
49
estimated
Long-horizon autonomy
54
estimated
Knowledge-work agents
36
estimated
Computer use
34
estimated
Security & context learning
46
estimated

Price

Input
$1.00/M
Output
$2.00/M
Cache read
$0.20/M
3:1 blend
$1.25/M
Agent-effective
$0.40/M

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

Agentic benchmarks

BenchmarkScoreNormalisedIndex weightHarness / source
SWE-bench Verifiedest.8%predicted 76
Terminal-Bench 2.0est.8%predicted 67
DeepSWEest.6%predicted 45
FrontierSWEest.5%predicted 13
CursorBenchest.4%predicted 36
GSO-Benchest.4%predicted 25
METR Time Horizonsest.10%predicted 74
Vending-Bench 2est.6%predicted 42
PostTrainBenchest.4%predicted 24
APEX-Agentsest.8%predicted 42
GDPvalest.7%predicted 45
Remote Labor Indexest.5%predicted 3
DeepResearch Benchest.5%predicted 48
OSWorld 2.0est.6%predicted 8
OSWorldest.4%predicted 72
Cybenchest.5%predicted 71
CL-benchest.5%predicted 21

Other benchmarks

BenchmarkCategoryScoreSource
CritPtReasoning9.1%source
SciCodeCoding50.2%source