I hand-wrote a 500-LoC RL stack to make hacking on RL research much easier.
Most RL stacks are either massive and unhackable, or duct-taped research scripts. I am open-sourcing Mithrl, a modular RLVR stack.
Next items on my checklist: adding more complex environment examples, supporting multi-gpu + async RL, and QoL fixes.
I might scrap external runtime dependencies (Huggingface PEFT + vLLM) and write purpose-built, simpler versions from scratch if I feel the need.
If you want to experiment with RL and are looking to own sovereign tools, Iโd love to get on call, understand your requirements and help integrate for free.
Tinker changed how people do post-training. We're big fans, and so are many of the teams we work with. But as those teams scale up, they keep raising the same two problems: per-token pricing gets expensive quickly, and youโre limited to the models and features the provider decides to support.
We think a good API deserves more than one implementation.
Today we're releasing Spindle, an open-source, Tinker-compatible training server that runs in your Modal workspace. Your existing Tinker scripts run unchanged, and because you own the server, you pay for GPU time instead of tokens.
I'm looking for a founding engineer [ SWE ] for Ambition.
If you're an avid tinkerer, love learning new things and extending your capabilities, I'd love to talk to you. Pay and equity is top of market with a lot of upside. Looking for people with deep technical curiosity and a history of ramping up on new domains and going deep.
Needless to say, your Ambition card will be issued on day 1.
Kimi K3 is live on Modal.
Moonshot has shipped the world's first open 3T-class model, and we're a day zero launch partner.
We trained a custom DFlash speculator for K3's novel architecture so you can run it faster, losslessly. The most capable open model we've worked with by far.
Should we be worried about how good AI is getting at coding autonomous drones?
Introducing Drone-Bench, a benchmark where AI agents code drones to complete a simple autonomous surveillance task.
Drone-Bench is independent but based on Project Pilot, our work with Anthropic.
Inkling by @thinkymachines is now available on Modal, backed by a custom DFlash speculator for 67% higher throughput and interactivity.
Running on Modal Auto Endpoints with SGLang today.