Thinking Machines asks users if Tinker needs training beyond 256k context

A Thinking Machines-affiliated researcher is surveying users on X about how valuable it would be to train models with contexts beyond 256k tokens on the Tinker fine-tuning platform, and how long such contexts should be.

2026-11-09 ~ 2026-11-09 · 2 related posts

• Agent researcher asks: how valuable is training beyond 256k context on Tinker? — cHHillee · 2026-11-09
• Thinking Machines polls users on whether Tinker training contexts beyond 256k are worth it — cHHillee · 2026-11-09