Hardware
we burned through 5 billion tokens in 48h. turns out giving everyone unlimited access to the most expensive model on the planet is not a gre…
we burned through 5 billion tokens in 48h. turns out giving everyone unlimited access to the most expensive model on the planet is not a great business strategy 🙈 So now we have 2 free opus sessions/d
we burned through 5 billion tokens in 48h. turns out giving everyone unlimited access to the most expensive model on the planet is not a great business strategy 🙈 So now we have 2 free opus sessions/day. if you blow through them, you can give Kimi, MiniMax, or GLM a spin, they're genuinely good. want more Opus? HF Pro is $9/mo. Way cheaper than the @Anthropic subscriptions and you get GPU access, Inference provider credits and a lot of other goodies. basically free given what Opus costs Anthropic to run not gatekeeping, we're just not made of money. 15k a day is crazy. Introducing ml-intern, the agent that just automated the post-training team @huggingface It's an open-source implementation of the real research loop that our ML researchers do every day. You give it a prompt, it researches papers, goes through citations, implements ideas in GPU …
Related
- 5B tokens in ml-intern in 48h 😅😅😅
- I’ve been using ml-intern for a while, and it genuinely changed my workflow. It's super good at: - Model/Dataset discovery. - Post-Training …
- This is crazy. ml-intern just passed the @huggingface internship test in 15 minutes. The task: replicate a research baseline from a DeepMind…
- Karpathy's autoresearch repo started an impressive trend. Agents can now train AI models to build SoTA agentic systems. And to think this is…
Source: Clem Delangue (X) | 2026-04-23