A new 594B-parameter mixture-of-expertsA model built from many specialist sub-networks where only a few activate per token, giving big-model capability at small-model running cost.Full definition → (17B active) with image input and adjustable effort is available for sampling and fine-tuningTaking a trained model and training it a bit more on your own examples so it gets better at one specific job.Full definition → on Tinker today. Apache 2.0 weights are promised for next week after external red-teaming.
Terms in this piece · Glossary
multimodal — A model that works with more than text — reading images, audio, or video, and sometimes generating them too.
mixture-of-experts — A model built from many specialist sub-networks where only a few activate per token, giving big-model capability at small-model running cost.
fine-tuning — Taking a trained model and training it a bit more on your own examples so it gets better at one specific job.
articleInkling (Thinking Machines Lab)Latent Space
articleThinking Machines introduces Inkling, an open-weight 975B MoE modelThinking Machines
articleInkling Small Lands Within A Point Of Inkling On The Artificial Analysis Intelligence Index With Less Than A Third Of The ParametersArtificial Analysis articles