Benchmark·Public sample

CoreCraft

A simulated enterprise environment where agents complete customer-support and operational tasks inside a fictional PC retailer.

Company-reportedConfidence: high·Last verified: 2026-08-16·How verification works
Access
Public / Open
Profile completeness 100%

What it is

CoreCraft models a company with customers, orders, support tickets, records, and tools. Surge reports evaluating nine frontier models on 150 workplace tasks ranging from lookups to multi-step operational workflows.

Capabilities

Only evidenced capabilities are shown as Yes. Missing evidence remains Unknown.

StatefulYes
Long horizonYes
Multi-turnUnknown
Multi-agentUnknown
AdversarialUnknown
MultimodalUnknown
Tool useYes
Browser useUnknown
Computer useUnknown
Code executionUnknown
File manipulationUnknown
API interactionUnknown
MCP supportUnknown
SandboxedUnknown
Deterministic resetUnknown
Stochastic behaviorUnknown
Human in the loopUnknown
Expert authoredYes
SyntheticUnknown
Production derivedUnknown
Real application replicaNo - evidenced
Partial observabilityUnknown
Persistent stateUnknown
Cross-applicationUnknown

Sources & provenance

  1. Surge AI Research Team · official · accessed 2026-08-16 · company-reported
  2. Surge AI · official · accessed 2026-08-16 · company-reported

Related artifacts