SShogo Bench

Claude-Sonnet-3.5

Balanced Claude model for coding, analysis, agent workflows, and cost control

Shogo Agent Index

Estimated#89 of 92
14.8range 10.9–18.8 · 19% of weight measured
Agentic coding
14
11% measured
Long-horizon autonomy
25
50% measured
Knowledge-work agents
14
estimated
Computer use
4
estimated
Security & context learning
14
50% measured

Price

Input
$2.60/M
Output
$13.00/M
Cache read
$0.26/M
3:1 blend
$5.20/M
Agent-effective
$1.20/M

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

Agentic benchmarks

BenchmarkScoreNormalisedIndex weightHarness / source
SWE-bench Verifiedest.8%predicted 42
Terminal-Bench 2.0est.8%predicted 4
DeepSWEest.6%predicted 6
FrontierSWEest.5%predicted 0
CursorBenchest.4%predicted 14
GSO-Bench4.6%4%OpenHands
METR Time Horizons45.2%10%source
Vending-Bench 2est.6%predicted 1
PostTrainBenchest.4%predicted 9
APEX-Agentsest.8%predicted 7
GDPvalest.7%predicted 12
Remote Labor Indexest.5%predicted 0
DeepResearch Benchest.5%predicted 39
OSWorld 2.0est.6%predicted 0
OSWorldest.4%predicted 8
Cybench17.5%5%source
CL-benchest.5%predicted 11

Other benchmarks

BenchmarkCategoryScoreSource
Aider PolyglotCoding51.6%source
BALROGLong-horizon32.6%source
DTBenchKnowledge67.8%source
GeoBenchMultimodal62.0%source
GPQA DiamondReasoning55.3%source
Humanity's Last ExamReasoning4.1%source
MATH Level 5Math56.9%source
MMLULegacy87.3%source
OTIS Mock AIMEMath8.5%source
SimpleBenchReasoning41.4%source
TheAgentCompanyKnowledge work24.0%source
VPCTMultimodal33.0%source