Research @thinkymachines | Previous: multimodal, visual coding @GoogleDeepMind
Mountain View
Joined May 2019
- Tweets391
- Following1.6K
- Followers3.5K
- Likes1.9K
Mu Cai retweeted
We are making lots of improvements to Tinker. Now we have a new optimizer to try out!
Dimuon is a new optimizer we built for LoRA fine-tuning. In our SFT tests, it often reaches lower loss in fewer steps than AdamW. We're still experimenting with Dimuon; you can Tinker with it today and share your feedback!
See our optimizers guide at tinker-docs.thinkingmachines…
Mu Cai retweeted
Tinkerers have been busy scaling up long-context RL! We’ve made significant improvements to Tinker’s efficiency to support those, and are passing these on with price cuts up to 70%.
GLM-5.3-Flash and DeepSeek-v4.1-Flash are also live for cost-efficient long-context work.
Mu Cai retweeted
This isn't the most notable aspect of today's news, but on the user data issue, there are different kinds of *training on user data* with very different privacy/IP implications. Sadly, AI cos don't like to disclose what they're doing.
- pretrain on user data, with users' tokens as prediction targets: high regurgitation risk, improper
- use user prompts to distill large models into small ones: low regurg. risk, some companies probably do this
- use user traces to construct RL tasks: low regurg. risk, because RL has low memorization abilities, but can extract customer IP, depending on how it's done. Ranges from benign "use explicit user feedback in reward model training" to invasive "upload user's coding environment and commit history to turn into rl envs"
"De-identification" is weak -- you can identify someone with a small number of bits, and long traces have more than enough. And it doesn't affect IP leakage concerns.
Mu Cai retweeted
Tomorrow will be my last day at Google after 27 years, and watching it grow from 25 people to 190,000+ has been an amazing journey. Below is a note I shared with many people internally at Google today. An excerpt is:
It has been an absolute pleasure to work with you and to help build some of the most widely used and impactful products of all time. As a kid, I dreamed of helping build software that would be used by many people, and Google now has thirteen products used by more than a billion people (amazing!). Our work has had a tremendous impact in the world, and I have been lucky enough to collaborate and form friendships with many colleagues that I deeply admire, respect, and enjoy. It still brings me joy every time I see people out in the world using our products to find information, handle email, translate documents, watch videos, learn new things, navigate and understand the physical world, browse the web, use their phone, run large-scale computations on our infrastructure, ride in an autonomous vehicle, or perform complex tasks with the help of our AI systems. I hope you all share this sense of joy, because it is a shared accomplishment! Thank you to all of my colleagues at Google over many years!
Now I'm excited to go start @DiscoLoopAI with my longtime friends and colleagues @Sanjay_Ghemawat, @OriolVinyalsML, and @quocleix.
(Updated post: slightly redacted to not have some personal info)
Mu Cai retweeted
We love open weights and plan to keep releasing open-weight models and fine-tuning tools. But we’re not absolutists; misuse risks are real. Here’s how we’re thinking about a safe path forward, and the research needed to get there. Come work on it with us.
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.
thinkingmachines.ai/blog/a-s…
🚀small but strong!
Try finetuning Inkling-small!
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.
thinkingmachines.ai/news/ink…
Fine-tune it on Tinker today, or chat with it in text, image, and audio on Tinker Playground.
Mu Cai retweeted
Inkling-Small is comparable to Inkling at a quarter the size. Weights are open, fine-tunable on Tinker today. Look forward to seeing what people make with it.
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.
thinkingmachines.ai/news/ink…
Fine-tune it on Tinker today, or chat with it in text, image, and audio on Tinker Playground.
Mu Cai retweeted
Presenting my grand unified theory of ML researcher impact: Your impact is directly proportional to how much pain you cause to infra.
Fundamentally, you can only inflict pain upon infra if your approach actually works. And the better your approach works the more pain infra is forced to endure.
So, to give some examples:
- MoE's add a ton of data-dependent computation => pain (shazeer++)
- GDN/KDA are the most complex architecture I've been forced to care about and a very annoying matrix inversion => pain (sonta++)
- Muon is much more annoying than Adam and causes annoying restrictions on parallelism => pain (keller/jeremy++)
- RL scaling forced many researchers to care about LLM inference and RL infra as a category => pain (tworek++)
Even papers like Attention Is All You Need have lead to significant pain! Before transformers were invented everyone was running small jobs and I never needed to think about kv-caches or 6D parallelism.
Mu Cai retweeted
The knowledge that makes AI useful is diffused. It lives with scientists, engineers, clinicians, firms. For AI to benefit from distributed knowledge, it must itself be distributed. Agree with Jensen that this is a future worth building.
For my first post, I’m sharing a letter @NVIDIA signed on why open models matter.
AI will transform every industry, power every company, and be built by every country.
Open models strengthen safety and cybersecurity, accelerate innovation and diffusion, and enable sovereignty.
The world needs both frontier closed models and frontier open models.
images.nvidia.com/pdf/Open-W…
Mu Cai retweeted
Inkling from @thinkymachines on ARC-AGI (Verified)
- ARC-AGI-2: 36.5%, $0.64/task
- ARC-AGI-1: 79.5%, $0.30/task
As of today, Inkling is the highest-scoring open-weight model evaluated by ARC Prize on both ARC-AGI-1 and ARC-AGI-2.
Mu Cai retweeted
Built a podcast clipping app with Inkling from @thinkymachines ✨
The model is exceptional at reasoning over long-form audio - so I have it listen to full episodes and direct FFMPEG on which clips to cut.
You can have it choose the best moments or search for specific topics 👇
Our Inkling is:
Native multimodal encoding
Pretraining from scratch
Owning strong visual coding capability
So proud of the team!
Today, we are introducing Inkling.
Inkling reasons efficiently across text, image, and audio modalities. We are making the full weights available.
thinkingmachines.ai/news/int…
Available today for fine-tuning on Tinker. Play with it in the Inkling Playground. 🧵
Mu Cai retweeted
Inkling is out today, with open weights and in Tinker. It's been fun to watch this one come together: pretraining began last winter, and starting in mid-January a small team built up the coding, reasoning, and agentic training from there. We learned a lot building it, and I hope people find good uses for it.
Today, we are introducing Inkling.
Inkling reasons efficiently across text, image, and audio modalities. We are making the full weights available.
thinkingmachines.ai/news/int…
Available today for fine-tuning on Tinker. Play with it in the Inkling Playground. 🧵
Mu Cai retweeted
🚨 Your Embedding Model is SMARTer Than You Think! Single-vector models actually hide powerful multi-vector capabilities in their frozen hidden states. We introduce SMART, a framework that unlocks this ability for SoTA multimodal retrieval. 🧵👇 🔗 huggingface.co/papers/2605.2…
Mu Cai retweeted
Collaborative AI runs on interactivity: machines and people, working in real time, across every modality. Solving it takes a community, join us.
We are offering grants of $100,000 + Tinker credits to researchers advancing the field of human-AI interactivity. Submit your proposals by June 19th!
thinkingmachines.ai/news/int…
Wow, always high quality papers from Xueyan and Yuheng, could be a good measure for video generation!
🔥Excited to share the first released work from our IEI lab! Congrats to @AnteaWu 🎉
This work is motivated by the lack of quantitative evaluation for physics alignment in video world models. With tools like MegaSam and CoTracker, we can directly reconstruct dynamic 3D scenes, enabling quantitative evaluation of physical alignment.
Both code and data are released — feel free to try it out! It should work, but if it doesn’t, contact @AnteaWu directly : )
Call for high quality realtime video/audio full duplex evals! The whole field needs them!
Come submit here!
We are offering grants of $100,000 + Tinker credits to researchers advancing the field of human-AI interactivity. Submit your proposals by June 19th!
thinkingmachines.ai/news/int…
Mu Cai retweeted
We are offering grants of $100,000 + Tinker credits to researchers advancing the field of human-AI interactivity. Submit your proposals by June 19th!
thinkingmachines.ai/news/int…
Mu Cai retweeted
We are so back!
People talk, listen, watch, think, and collaborate at the same time, in real time. We've designed an AI that works with people the same way.
We share our approach, early results, and a quick look at our model in action.
thinkingmachines.ai/blog/int…