Thinking Machines releases Inkling 2 Small, weights due next week
The 594B-parameter mixture-of-experts model is on Tinker now. The lab plans Apache 2.0 weights after safety testing.
Thinking Machines Lab
Every ClaudeGPT Si story about Thinking Machines Lab, newest first: news, analysis, reviews and guides.
2 stories · Updated
ResearchExplainer
What is mixture of experts? A 594B model that uses 17B per token
MoE models store hundreds of billions of parameters but run only a few percent of them per token. Here is how that works, from the papers to Inkling 2 Small and Kimi K3.1 Pro.