docs: add Modal GPU serving guide - #424
Open
felipeocampoos wants to merge 1 commit into
Open
Conversation
How we used Modal as a drop-in replacement for MIT SLURM GPU partitions during the BODHI-Medcalc rerun campaign (Sept 2026), when MIT's GPU QOS was maxed out. Covers the one-class-per-model vLLM serving pattern and the seven gotchas that cost real time: silent GPU-quota exhaustion without max_containers, TP=2+AWQ hangs, incomplete client venvs, the checkpoint-hash relaunch trap, per-model SLURM time limits, HF Hub rate-limiting on concurrent dataset loads, and NFS cache races. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
max_containers), TP=2+AWQ hangs, incomplete client venvs, the checkpoint-hash relaunch trap, per-model SLURM time limits, HF Hub rate-limiting, NFS cache racesWhy
Sebastian asked for a shareable writeup so others on the team can use Modal when the MIT GPU queue is full, without rediscovering the same gotchas.
🤖 Generated with Claude Code