ComfyUI-ZeroCopy-Continuity kills VRAM reloads: 35.5× faster inter-shot latency
LatentForge-Dev · reddit · 2026-10-21
An open-source ComfyUI custom node suite, ComfyUI-ZeroCopy-Continuity, targets multi-shot DiT video pipelines (MiniMax-H3, Wan 2.1, HunyuanVideo, FLUX.1):
- VRAM pinning: ZeroCopyVRAMPin keeps 15-20GB DiT weights resident in VRAM via monkeypatched eviction and expandablesegments: True, cutting model reload from 7.8s to 0.0s.
- Lossless latent chaining: ZeroCopyLatentStore captures output latents in a resident GPU ring buffer; ZeroCopyLatentHandoff injects the previous shot's terminal latent directly into the next shot's conditioning — zero VAE decode/encode roundtrip, bit-identical continuity.
Benchmarks on RTX 5090: inter-shot overhead 14.2s → 0.4s (35.5×), reconstruction MSE 0.042 → 0.0, zero fragmentation after 20 shots. Available via ComfyUI Manager or git clone, with full example workflows in the repo.
More from Infra
- IFM open-sources RL360: 8,192 concurrent sandboxes and ~1.3M rollouts for agentic RL — rupspace · 2026-10-23
- SpaceX seeks access to 80,000 US cell tower sites; Polymarket puts carrier odds at 29% — Polymarket · 2026-10-23
- Modular open-sources MAX framework to free AI compute from single-vendor lock-in — clattner_llvm · 2026-10-23
- Claude Models on Google Cloud (Anthropic managed) is now generally available — claudeai · 2026-10-23
- Scaling GLM-5.3-Flash training to 1M token context, 64x over Megatron baseline — AndrewLBeam · 2026-10-23
- Perplexity's Portable Computer lands on AMD Ryzen AI Max, running local 27B model — AravSrinivas · 2026-10-23