EnterpriseBench: CoreCraft (Surge) (EnterpriseBench-CoreCraft)
We show this table for reference; we do not rank on it.
An enterprise-operations agent benchmark from Surge AI set in a simulated company (CoreCraft), where agents complete multi-step business tasks across internal tools and records.
Benchmark score on EnterpriseBench-CoreCraft — November 7, 2026
We compile the EnterpriseBench-CoreCraft rows from provider self-reports. Inkling 2 Small leads the table at 71.8%. We do not use these results to rank models overall.
1 modelAgenticCurrentDisplay onlyUpdated November 7, 2026
Benchmark score table (1 model)
| Rank | Model / configuration | Score | Parameters (B) | Open / closed |
|---|---|---|---|---|
| 1 | Inkling 2 SmallThinking Machines Lab | 71.8% | Not reported | Closed |
How to read this leaderboard
Operator receipt: 1 sourced row is currently displayable on this page; the leading published row is Inkling 2 Small at 71.8%.
Honest limit: Private environment; harness details are not published in the launch post.
About EnterpriseBench-CoreCraft
Year
2026
Tasks
Enterprise operations tasks in a simulated company
Format
Task success (%)
Difficulty
Multi-step enterprise tool use
Thinking Machines attributes EnterpriseBench to Surge and reports CoreCraft scores in the Inkling 2 Small launch (November 5, 2026): Inkling 2 Small 71.8%, Gemini 3.8 Flash 58.5%, GLM 5.3 51.8%, GLM-5.3-Flash 46.7% and Kimi K3 43.1%.
Freshness and provenance
Version
EnterpriseBench-CoreCraft 2026
Refresh cadence
Quarterly
Staleness state
Current
Question availability
Public benchmark set
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
Questions
What does EnterpriseBench-CoreCraft measure?
An enterprise-operations agent benchmark from Surge AI set in a simulated company (CoreCraft), where agents complete multi-step business tasks across internal tools and records.
Which model scores highest on EnterpriseBench-CoreCraft?
Inkling 2 Small by Thinking Machines Lab currently leads with a score of 71.8% on EnterpriseBench-CoreCraft.
How many models are evaluated on EnterpriseBench-CoreCraft?
1 AI models have published results on EnterpriseBench-CoreCraft in the BenchLM catalog.
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 6,100+ subscribers.
One email each week. Unsubscribe anytime.