Chakra Labs
151 posts
High-fidelity trajectories and environments for frontier AI research
- Thanks for featuring our work @immutablejacob, @MessariCrypto, and team! Better datasets & environments => better modelsDespite the incoming drought of public text, we're nowhere close to using up the world's data. The world's total data is ~200 ZBs and doubles every ~2 years. The valuable modalities and sources that can be drawn from include: > Audio (@psdnai, @silencioNetwork, @Meet_Perle) >
- Replying to @chakra_aiEx: access our Amazon clone using familiar Gymnasium style APIs.
00:00To get started, check out our docs and code examples below. - Chakra Labs repostedYou can now run tasks in @chakra_ai's RL environments using Browserbase as the runtime. Get task concurrency, isolated contexts, and enhanced observability.Update: Dojo now supports @browserbase
00:00 - Replying to @chakra_aiWe're excited to announce support for Browserbase, which allows researchers to quickly spin up 1000's of browser tasks concurrently using their serverless hosted infrastructure. Researchers now benefit from: - Hyper-scalable browser infra - Secure SOC II instances - NativeTo get started with this custom run-time, check out the Dojo engine docs below:
- Chakra Labs repostedthese ppl just keep cooking 🔥Another fun Dojo we published: Gmail We have a first party clone of Gmail inbox + email sending available free to use - with verifiable state and tasks
00:00 - Chakra Labs repostedAn interesting design decisions we made with dojo: - All of our dojos inherit the following state pattern
- Chakra Labs repostedso we did this :)
00:00if i was running Linear, and i wanted every AI ever to become very good at using Linear, i would make an RL environment version of Linear and open-source it for free, with a curated set of realistic sample data. every lab would train with it. who else can compete with that? - Chakra Labs repostedThis is awesome... This is how we craft the future of CUAs, by actually open-sourcing the RL env suites...Today, we’re excited to announce Dojo, a collaborative RL environment suite for computer use agents (CUA).
00:00


















