Inkling release paves the way for next-gen models

jonlachman · x · 2026-07-16

According to the reshared post, team members stated that while they are proud of what the team achieved with the release of Inkling, they value what they learned during the building process even more. Furthermore, the pretraining and infrastructure foundation from this release is already supporting their next generation of models.

As previously announced, Inkling is a multimodal model designed for text, image, and audio reasoning, with fully open weights, and is immediately available for fine-tuning on Tinker.

Related event: Thinking Machines launches Inkling, its first open-weight multimodal model(169 posts)→

Original post →

More from Multimodal

Multimodal channel →