{"slug": "runsnack-p2p-gpu-sharing-with-a-link-would-love-feedback-from-ml-people-who-rent", "title": "RunSnack – P2P GPU sharing with a link. Would love feedback from ML people who rent/rent out GPUs", "summary": "RunSnack, a peer-to-peer GPU sharing platform, lets users rent or share GPUs via a single link, with no accounts, cloud, or platform fees. The tool, built for quick ML tasks, uses sandboxed Docker containers and WebRTC for encrypted, P2P connections, and is seeking feedback from the Hugging Face community on security, pricing, and features.", "body_md": "Hey HF community,\n\nI built a tool that might be relevant to a lot of people here who either have idle GPUs or need compute for quick jobs. It’s called [RunSnack](https://runsnack.com/) — share your GPU with a single link. No accounts, no cloud, no platform fees.\n\n**Why this matters for ML people:**\n\n-\nNeed an A100 for 2 hours of inference testing? Find a host, get a link, instant terminal.\n\n-\nGot a 4090 sitting idle overnight? Share it, set your price, negotiate directly.\n\n-\nNo KYC, no cloud markup, no usage logs. Just P2P compute.\n\n**How it works:**\n\n-\nHost runs a script → gets a shareable link\n\n-\nRenter opens link in browser → instant xterm.js terminal with GPU access\n\n-\nEverything runs inside a sandboxed Docker container (read-only, privilege-dropped)\n\n-\nWebRTC data channel — encrypted, P2P, no central server sees your data\n\n-\nMarketplace is organic: post on Discord what you need or offer, negotiate, share link\n\n**Built for the “quick GPU snack” use case:**\n\n-\nML researcher needs 30 min of inference on a GPU they don’t own\n\n-\nStudent wants to fine-tune a model for a few hours without spinning up a cloud instance\n\n-\nSomeone with a homelab GPU wants to monetize idle time\n\n**I’d love feedback from this community on:**\n\n-\n**Would you use this?** Why or why not?\n\n-\n**Security concerns?** Docker sandbox + GPU passthrough — what would make you trust it enough to either host or rent?\n\n-\n**What’s missing?** Pre-installed ML frameworks? Persistent storage? Something else?\n\n-\n**Pricing expectations?** What would you pay per hour for, say, a 4090 vs an Jetson Nano via a P2P link like this?\n\n-\nIs this approach a competitor for sites like **RunPod**?\n\nThank you!\n\nLastly, I add this 30-sec YouTube reel so everyone gets better the idea", "url": "https://wpnews.pro/news/runsnack-p2p-gpu-sharing-with-a-link-would-love-feedback-from-ml-people-who-rent", "canonical_source": "https://discuss.huggingface.co/t/runsnack-p2p-gpu-sharing-with-a-link-would-love-feedback-from-ml-people-who-rent-rent-out-gpus/178525#post_4", "published_at": "2026-08-13 05:23:11+00:00", "updated_at": "2026-08-13 10:44:08.489029+00:00", "lang": "en", "topics": ["ai-infrastructure", "ai-tools", "ai-startups"], "entities": ["RunSnack", "Hugging Face", "RunPod", "Docker", "WebRTC", "A100", "4090", "Jetson Nano"], "alternates": {"html": "https://wpnews.pro/news/runsnack-p2p-gpu-sharing-with-a-link-would-love-feedback-from-ml-people-who-rent", "markdown": "https://wpnews.pro/news/runsnack-p2p-gpu-sharing-with-a-link-would-love-feedback-from-ml-people-who-rent.md", "text": "https://wpnews.pro/news/runsnack-p2p-gpu-sharing-with-a-link-would-love-feedback-from-ml-people-who-rent.txt", "jsonld": "https://wpnews.pro/news/runsnack-p2p-gpu-sharing-with-a-link-would-love-feedback-from-ml-people-who-rent.jsonld"}}