Laguna S 2.1 Collection Our most capable model to date, designed for long-horizon work. • 12 items • Updated 8 days ago • 36
RAGU: A Multi-Step GraphRAG Engine with a Compact Domain-Adapted LLM Paper • 2607.11683 • Published 17 days ago • 147
Harness Handbook: Making Evolving Agent Harnesses Readable,Navigable, and Editable Paper • 2607.13285 • Published 16 days ago • 227
view article Article Welcome Inkling by Thinking Machines +2 burtenshaw, merve, pcuenq, ariG23498 • 15 days ago • 138
Long-Horizon-Terminal-Bench: Testing the Limits of Agents on Long-Horizon Terminal Tasks with Dense Reward-Based Grading Paper • 2607.08964 • Published 21 days ago • 76
Agentic Abstention: Do Agents Know When to Stop Instead of Act? Paper • 2606.28733 • Published Jun 27 • 150
NatureBench: Can Coding Agents Match the Published SOTA of Nature-Family Papers? Paper • 2606.24530 • Published Jun 23 • 65