Inkling Open: Thinking Machines Lab's First Model Lands
Thinking Machines Lab has released Inkling, its first AI model, with co-founder and CEO Mira Murati framing the launch around new ways for people to interact with frontier systems. The news matters because Murati’s post-OpenAI lab is moving from talent story to shipped model, and it’s doing so with an open release.
What is Inkling from Thinking Machines Lab?
Inkling is the debut model from Thinking Machines Lab. That alone is the headline. The company has been watched closely because Murati, formerly OpenAI’s CTO, built the lab with heavyweight AI expectations attached.
The available announcement centers on interaction, not just raw model size. Murati’s stated aim is to bring new interaction models to the front. That suggests the company wants Inkling judged not only by benchmark charts, but by how people actually work with it.
Key details still matter. Public materials available here don’t include context-window size, pricing, licensing fine print, or benchmark scores. So the grounded read is simple: this is a first release, and the open-access angle is what’s drawing the most attention.
What does Inkling mean for open AI models?
This looks like a signal that frontier labs still see strategic value in openness, especially as developers push back against closed, API-only systems.
If Inkling is genuinely useful and genuinely open in practice, it could give researchers and builders another serious model to inspect, adapt, and compare. But without published scores or deployment terms in the material at hand, it’s too early to rank it against OpenAI, Anthropic, Google, or Meta systems.
Why is Mira Murati’s launch getting attention now?
Because this is the shift from biography to product. A former OpenAI CTO can attract attention by hiring. A first model has to earn it. Inkling is Thinking Machines Lab’s opening move.