AIThinking Machines Lab43m ago
Thinking Machines Lab debuts Inkling 2 Small, a MoE model
Thinking Machines Lab debuts Inkling 2 Small, a MoE model with 594B total and 17B active parameters, and plans to release weights next week after safety testing
TL;DRNew efficient AI model uses only 17B active parameters despite 594B total size.
Why it matters: Open-weight MoE models democratize access to capable AI; efficiency matters for deployment costs.
{"selector":"[data-shape]","layers":[{" nodes":[22, 33,24,44,35,46],"edges":[[22,33],[33, 44],[24,33],[24,35],[35,44],[35, 46]] …
Read full articleSource: Thinking Machines Lab · Opens in new tab