One-liner to grab 4x B200 GPUs: uvx modal shell --gpu b200:4
charles_irl · x · 2026-09-17
- The author shares a one-liner, uvx modal shell --gpu b200:4, to spin up a shell with 4 B200 GPUs via Modal, joking 'i just want to serve five petaflops'.
- It demonstrates combining uvx (the Python tool runner) with Modal's cloud GPUs: no local setup needed, one command gets you an interactive shell with the GPUs you specify.
- A directly copyable trick for developers needing on-demand heavy compute.
Related event: Anthropic Engineer's Compute Lament Highlights GPU Scramble(2 posts)→
More from Infra
- DeepSeek V4 Pro parsing bug said to hit ~60% of OpenRouter providers; fix upstreamed to sglang — michellechen · 2026-09-17
- Unconventional AI Open-Sources Un-0, an Image Generator Built on Coupled Oscillators — NaveenGRao · 2026-09-17
- Naveen Rao's Startup Taped Out Custom AI Chip in 5 Months, Demonstrates On-Chip Causal Dynamics — NaveenGRao · 2026-09-17
- Team tapes out physical-compute chip in 5 months, generates images for under 900 nJ each — NaveenGRao · 2026-09-17
- Which 'token brokers' give back to open source? New data ranks upstreamed PRs to OSS inference engines — michellechen · 2026-09-17
- XeBoostLM: native C++ local LLMs on Intel NPUs and iGPUs, zero Python — Spiritual-Ad-5916 · 2026-09-17