LaunchAI ModelsNovember 6, 2026

Thinking Machines releases Inkling 2 Small, 17B-active MoE

Read original source →latent.space

Inkling 2 Small is a 594B-total/17B-active multimodal MoE pretrained from scratch in under 3 days on <12T tokens in native NVFP4, with a single joint RL run. It scored 31 on the Artificial Analysis Intelligence Index, and Apache 2.0 weights are promised "next week" after external safety testing.

1 source

More stories today

Open the live feed