PewDiePie is doing his own GRPO training, and HF's merve wants to help but can't reach him
mervenoyann · x · 2026-10-01
Hugging Face researcher merve shares that PewDiePie is running his own GRPO (agentic RL) training, though he doesn't name his tooling. Per her notes: he tried distilling Sol but got banned a few times, built a website asking people to donate traces (nobody did — she notes he could just use the Hub), and she wants to help but can't reach him.
Related event: PewDiePie Trains His Own Local LLM with GRPO, Community Rushes to Help(4 posts)→
More from Fun
- Anthropic's IPO lands on the same day as GTA 6's release — firstadopter · 2026-11-16
- AI progress doubled since GTA VI's first trailer; GPT-4 Turbo now ranks 379th — ArtificialAnlys · 2026-11-16
- Dev burns entire Cursor quota to ship Ship Quest, a vibe coder comedy point-and-click — chongdashu · 2026-11-16
- Author confirms the Ricursive blog post was AI slop gone off the rails — natolambert · 2026-11-16
- He skipped the Switch: Claude + Blender MCP turns a phone into a motion-controlled game console — stablequan · 2026-11-16
- Nathan Lambert flags AI2 Ricursive blog paragraph as AI-generated slop — natolambert · 2026-11-16