cloud-worker

Tooling for running the MiniMax H3 open-weights video model (native stereo audio, 24 fps, 5–15 s, ~1 MP native) on RunPod GPU machines via ComfyUI.

Current test rig: 1x B200 (180 GB) pod — connection details in pod-info.txt.

Layout

• comfyui/ — ComfyUI configs stored for easy rebuild
  • extra_model_paths.yaml — installed to /ComfyUI/extra_model_paths.yaml
  • workflows/ — the three MiniMax H3 workflows (t2v, i2v, r2v), installed to /ComfyUI/user/default/workflows/
• scripts/pod/ — runs on the pod
  • install-minimax-h3.sh — one-shot rebuild (ComfyUI clone + deps + configs
    • workflows + images + weight downloads + launch); idempotent, survives pod resets since it stages from /workspace
  • download-weights.sh — all weight families → /workspace/minimax-h3/models
  • start-comfyui.sh — headless ComfyUI in tmux on 127.0.0.1:8188
  • make-test-images.py — synthetic reference/first-frame images → /ComfyUI/input
• scripts/local/push-to-pod.sh — stage repo files onto the pod's persistent volume
• scripts/local/port-forward.sh — persistent tunnel localhost:8188 → pod ComfyUI panel (auto-reconnects)
• bench/minimax_bench.py — API-based benchmark harness (t2v / i2v / ref2v × duration × resolution × weight family); appends to bench/results.csv + bench/results.jsonl
• docs/WEIGHTS.md — which weights are downloaded, where, and which family
• docs/RESEARCH.md — model/ComfyUI findings, running without ComfyUI (SGLang / vLLM-Omni / diffusers), concurrency model, RunPod serverless
• docs/BENCHMARKS.md — measured results on the B200
• docs/CONCURRENCY.md — concurrency model + cost-per-video math across GPU tiers
• docs/MITIGATIONS.md — per-machine-class OOM map and required ComfyUI flags/patches
• docs/SERVING-PLAN.md — paid/free pool design, pricing tables, capacity & demand scenarios

Quick start

# rebuild the pod from scratch (e.g. after a reset wiped the container disk)
scripts/local/push-to-pod.sh
ssh -p 17561 root@216.243.220.136 'bash /workspace/minimax-h3/install-minimax-h3.sh'

# tunnel to the ComfyUI panel (leave running)
scripts/local/port-forward.sh &
open http://localhost:8188

# one-off generation through the API
python3 bench/minimax_bench.py --task t2v --width 864 --height 480 --seconds 5

# smoke suite / full sweep
python3 bench/minimax_bench.py --suite quick
python3 bench/minimax_bench.py --suite sweep --prefix prunedint8_

On the pod, ComfyUI runs in tmux session comfyui (tmux attach -t comfyui), logs at /workspace/minimax-h3/comfyui.log.