Thinking Machines Lab just launched their first product, Tinker Tinker is a distributed training API for fine-tuning language models 🔹Gives researchers & developers low-level control over algorithms and data while abstracting away the complexity of distributed training & deployment 🔹 Lets you fine-tune open-weight models like Llama and Qwen, including large MoE variants like Qwen3-235B-A22B The @Anyscale team had early access to experiment with the API and wrote up a simple end to end example

Tinker loves to travel, she even trailers herself!😄❤️#horse #horsepower #beautifulanimals

Hold on to your potatoes! Ke Huy Quan & Ginnifer Goodwin take on Indiana Jones Adventure #Disneyland

Disney Imagineering brings Olaf to life, will be seen at Disneyland Hong Kong and Paris

Kids Go To Disneyland🎡- R for Rabbit