ml-intern is fully on mobile now you can launch 8 A100s from your phone. while on the couch. while commuting. wherever I just did this while biking. same sessions as your desktop too — start a run on your laptop, check on it from your phone, it's all there... the research lab is now just wherever you areshow more

This is great I have been experimenting with using and accessing @Gradio Spaces in mobile applications.

mobile-launch for gpu uworkloads is quietly the underrated change this year. the always-on agent becomes a product!

Wow. I want to give this a try. What’s the best way to figure out cost structure for someone who isn’t AI native, but has a problem to work on? I imagine it could be expensive to test out 50 different training runs.

Stage1. Launch at desk. Stage2. Lanuch lying on sofa or bed. Stage3. Lanuch at dream or in your mind

@lvwerra daily major update on this product is so cool

This is the reason compute prices are spiking

Are you actually training on the HLE benchmark dataset itself, or is “HLE” just shorthand for training/evaluating on HLE-style tasks using separate data? I’m asking because training directly on HLE would contaminate the benchmark and make any resulting score hard to interpret.

There are open discussions in the space on hugging face. Can you please take a look

That’s good, was finding it difficult to resume a session from my pc to my mobile. Love it

@lvwerra What is the pricing on GPU sessions?

"Love the mobile accessibility, but what about integrating voice commands? Imagine launching A100s with voice assistants, streamlining the process even further. Would be a game-changer for researchers on-the-go"