README.md

---
license: mit
base_model: MiniMaxAI/MiniMax-H3
tags:
- comfyui
- workflow
- video
- audio
- text-to-video
- minimax-h3
- multishot
---

# MiniMax-H3 Multishot Workflow

Chain multiple MiniMax-H3 shots into one continuous video **with audio** - in
one node. Each shot starts from the last frame of the previous one; the
duplicated seam frame and its 1/24s of audio are trimmed automatically.

The demo below was made **by this workflow**: 30 seconds, three chained shots
from one script, same presenter and same voice across both seams, rendered on
the Q5_1 GGUF.

<video controls src="https://huggingface.co/joeygambino/MiniMax-H3-Multishot-Workflow/resolve/main/H3_multishot_presenter_demo.mp4"></video>

## Files

- **H3_Multishot_AIO.json** - easy mode: loaders > script box > one sampler
  node > save. Write one prompt per shot with `---` between them, pick a
  shot count (0 = one shot per prompt, 1-8 forces it), queue.
- **H3_Multishot_3chain_expert.json** - the same pipeline exploded into three
  visible shot chains for per-stage tinkering.

## Requirements

- ComfyUI **v0.30.0+** (native MiniMax H3 support)
- The node pack: [ComfyUI-H3-Multishot](https://github.com/jlucasmcrell/ComfyUI-H3-Multishot) (Manager > Install via Git URL;
  includes the one-line ComfyUI-GGUF architecture patch)
- Models: [MiniMax-H3 GGUF quants](https://huggingface.co/joeygambino/MiniMax-H3-GGUF) (Q5_1 for 24-32 GB cards, Q4_0
  for 16 GB) or the originals; text encoder + VAEs from
  [Comfy-Org/MiniMax-H3](https://huggingface.co/Comfy-Org/MiniMax-H3)

## Notes

- `frames_per_shot` sits on H3's 17k+5 frame grid (243 = ~10.1s at 24 fps;
  362 = ~15.1s, the trained max).
- End every shot on what the NEXT shot expects to see - the chain hands each
  shot the previous final frame, and matching that bridge to the next shot's
  framing is what makes seams invisible.
- Malformed JSON scripts fail loudly instead of rendering the raw text.

## v1.1 (2026-08-04)

- **Image-to-video.** The Multishot Sampler now takes an optional `start_image`.
  Connect a `LoadImage` and shot 1 starts from that frame, then chains as usual.
  Leave it unconnected for unchanged text-to-video behaviour - v1.0 graphs keep
  working.
- **~4x faster on 32GB cards.** The text encoder is evicted before sampling.
  The Qwen3-VL encoder (~16.5GB even at Q4) and the H3 DiT (~25GB) do not co-fit
  on 32GB, so the DiT was loading *partially* and streaming ~19GB from system RAM
  every step. Measured on an RTX 5090: **~60 min -> ~15 min** for the same render.
- **Text encoder GGUF + mmproj published:**
  [joeygambino/MiniMax-H3-encoder-GGUF](https://huggingface.co/joeygambino/MiniMax-H3-encoder-GGUF).
  The mmproj sidecar is **required for multi-shot**, not just for reference
  images - shot chaining feeds the previous frame through the encoder's vision path.
- Render at H3's native resolution and upscale afterwards: native 1920x1088
  measured *worse* than 960x544 in blind review and cost ~4x the time.

Node pack: [ComfyUI-H3-Multishot](https://github.com/jlucasmcrell/ComfyUI-H3-Multishot)

## Support

Everything here is free and stays free. If it saved you a night of debugging,
tips keep the 5090 warm:

* [Buy me a coffee on Ko-fi](https://ko-fi.com/joeygambino)
* [Sponsor on GitHub](https://github.com/sponsors/jlucasmcrell)
* [Liberapay](https://liberapay.com/joeygambino) (recurring)