• fine-tuning — Taking a trained model and training it a bit more on your own examples so it gets better at one specific job.

Specific credit amounts for academic fine-tuning work: $250 per student for coursework and grants from $5,000 for research projects or open-source software built on Tinker.

• postThinking Machines opens a research blog, starting with inference nondeterminism
• articleTinker is now generally available.
• postThinking Machines argues for staged access, using Inkling as the test case
• postOn-policy distillation: RL's error correction at SFT prices

• postTinker exposes distributed fine-tuning behind your own Python training loopThinking Machines
• articleTinker is now generally available.Thinking Machines
• articleScience is best shared!Thinking Machines

Checking sign-in…

Loading comments…