Thinking Machines
@thinkymachines
Thinking, beeping, and booping.
@tinkerapi
1 Following 183.7K Followers
Our own @johnschulman2 talks with Dwarkesh about where human judgment still matters as models improve and self-improve: teaching them to handle messy real-world tasks, applying taste to what works in the long run, and, above all, specifying what we actually want.
Show more
Cleaning data and aligning the reward function for RLVR takes expertise and effort upfront, but the result is a model that's state-of-the-art on a complex task.
Guest post by researchers at UIUC and Bridgewater, in collaboration with our team.
Show more
LLMs with scaffolds have lagged on text-to-SQL, a task that relies on human judgment. By folding expert judgment into every part of RLVR on Tinker, @maxYuxuanZhu and @ddkang (UIUC and Bridgwater) trained the first text-to-SQL model to beat the human mark.
Show more
Today, we are launching Tinker grants of up to $50,000 in credits for safety research on open-weight models. We share some project ideas that excite us below; if you’re working on a safety project that could be accelerated by additional Tinker credits, we want to hear from you!
Show more
Try out Inkling and Inkling-Small on OpenRouter here:
We want to improve Inkling’s agentic performance. To help us understand its real-world behavior, we are making it available for free on OpenRouter (only with agentic harnesses) for the next few weeks, starting now. We’ll use the data, disassociated from accounts, to better it.
Show more
Releasing weights indiscriminately isn't safe. Neither is keeping capable models inside a few labs.
We think there's a path between them. We haven't mapped all of it. Our new post covers the part we can see: how we assessed Inkling, and why access should widen in stages.
Show more
Another open-weight release from @thinkymachines 👀 Inkling-Small is here.
With native reasoning over audio and images and variable thinking effort, it's a great choice for fine-tuning, with NVIDIA NeMo on NVIDIA DGX Station.
NVFP4 checkpoint here:
Show more
Today, we are releasing Inkling-Small.
Inkling-Small achieves comparable performance to Inkling at a quarter of its size. It features 276B total parameters, 12B active. We are making the full weights available.
Fine-tune it on Tinker today, or chat with it in text, image, and audio on Tinker Playground.
Show more
Safety work is only as good as the access researchers and defenders have to real models, and open sharing is how the whole ecosystem gets stronger. Glad to support the Open Secure AI Alliance.
AI security advances when the industry builds in the open, together.
We're introducing the Open Secure AI Alliance with industry leaders to develop new techniques and tools to safeguard software and agents.
By sharing models, tooling and research in the open, we can broaden the community of defenders.
Learn more about the founding members’ contributions:
Show more
Today, we are introducing Inkling.
Inkling reasons efficiently across text, image, and audio modalities. We are making the full weights available.
Available today for fine-tuning on Tinker. Play with it in the Inkling Playground. 🧵
Show more
We're building AI that people and organizations can shape and make their own. AI should extend our will and judgment instead of neglecting it; enabling that is the technical challenge we are working to solve.
Show more
We are offering grants of $100,000 + Tinker credits to researchers advancing the field of human-AI interactivity. Submit your proposals by June 19th!
1. (System design) - The Interaction Models see your screen and collaborates with you live. Here we're building a scalable system architecture together — no copy-pasting, no switching tabs, just thinking out loud and drawing on the screen together.
Show more
The technical report includes our motivation, early evaluation results, and technical approach.
Lili and Martin get some help controlling themselves.
The model works in more real situations too, like when Mianna is trying to say words she's only read before.
and draws when things are easier to explain visually!
The model can multi-task!
Long thinks the model knows everything, but the model actually searched while listening and responding to him so he didn't notice.
The team has been sweeping at local trivia night thanks to a model that's aware of continuous time.
With the model's simultaneous speech capability, Horace has gotten a lot easier to work with recently.
