Benchmark family

ToolSandbox

A family groups the versions of one benchmark. It is not evaluated and has no applicable models of its own; each of its 1 definition is resolved separately.

Multi Turn AgentStateful AgentTool Use

Definitions

Newest first, following the recorded successor chain. Each definition is resolved against models separately.

  1. BenchmarkToolSandbox
    ActiveNewest in family
    Scale
    3.8 tools
    Primary metric
    Milestone Similarity
    Applicable models
    0

Family methodology

No family-level methodology recorded.

Across definitions

Tasks
Insufficient Information Handling, Stateful Conversational Tool Use, Tool Input Canonicalization, Tool State Dependency Reasoning
Sources
None recorded at family level
Last reviewed
2026-09-23

Applicability does not mean evaluated; benchmark result/model score data is outside this UI contract.