Small fix : audio reference will now pin across multiple hops
Major update
Hand Tie Clips 2.0 — Native MiniMax H3 Chains in ComfyUI
I highly suggest feeding the git/hf page to a LM to study all the respective settings for control within your planning itself. It has a LOT of fine tune controls available now. This is NOT a plug-and-play one click setup, requires some tinkering to achieve what you want.
MiniMax H3 produces stunning six-second clips, but building longer sequences is notoriously messy—character faces drift, lighting shifts, and audio restarts on every generation. Hand Tie Clips runs your entire sequence in a single queue: it passes final frames forward and locks character consistency from start to finish.
Key Features
🎙️ Continuous Voiceover: Drop in a single audio file. Every shot automatically lip-syncs to its own slice without audio resets, seam cuts, or manual re-timing.
🧍 Persistent Character Consistency: Feed it one reference photo. The node carries the identity through every shot, preventing face drift across the chain.
🎬 Beat-Based Directing: Write simple one-line actions per shot ("She turns to the window," "He walks to the counter"). The node handles prompt generation, camera movements, and cut transitions.
🔄 SWAP Mode: Combine motion from a video with identity from a photo—the source clip drives movement and scene context, while your reference photo drives the person.
✍️ Built-in Scripting (WRITE Tab): Connect a local LLM to automatically draft multi-shot sequences directly from a short concept paragraph.
Quick Start
Drop the
ComfyUI-Hand-Tie-Clipsfolder into yourcustom_nodes/directory and restart ComfyUI.Load the included starter workflow.
Add a character photo to the REFERENCES tab, write 3 shots, and hit queue.
Includes starter and showcase workflows out of the box, with detailed tooltips for every setting.
Needs
ComfyUI new enough to include Add Guide for MiniMax H3 (
MiniMaxH3AddGuide)A ref2va (or hybrid ref2va) checkpoint — fl2va has no reference rows
Video VAE + audio VAE + MiniMax text encoder
Recommended: ComfyUI-H3-Motion-Context for the latent join. Without it the node falls back to
MiniMaxH3AddGuide, which still works and still chains — it is a different join, so a seam you are comparing against someone else's render may not be the same code path.
The two shipped workflows wire the turbo stack this node is actually run with, because an example without it is not the graph anyone uses:
UNETLoader -> LoRA Loader Stack -> H3 AdaLN LoRA Fix -> MiniMax H3 Low VRAM
-> H3 SLA Attention -> Model Preview Override -> Hand Tie Clips
packnodes inside workflow :
ComfyUI-PlagueKind-NodesLoRA Loader Stack, H3 AdaLN LoRA Fix, H3 SLA Attention
ComfyUI-KJNodesMiniMax H3 Low VRAM Attention (experimental), Model Preview Override
Realistic Limitations
⚠️ Quality Drift: Image quality degrades on long chains. 3 to 5 shots is the realistic ceiling before colors and textures begin to wander.
📊 Degradation Study: The README details a 10-run, 9-shot study measuring exact failure points and identifying settings that yielded no impact.
Requirements & Links
ComfyUI Build:
v0.34.0or newer (required for H3 support)Dependencies: None (no extra Python packages needed)
🔗 GitHub Repository & Full Docs: https://huggingface.co/sandpies/ComfyUI-Hand-Tie-Clips
Description
V1.1
Hand Tie Clips 1.1.0
The fix worth updating for: a reference pinned to a later hop tended to be rendered as the shot. The line that tells the encoder what a picture is for was only emitted on hop 1, so a still scheduled onto hop 3 arrived uncited — and a wardrobe plate could be announced as the character's face. Fixed.
- The panel is four tabs now — Script / Refs / Media / Write — with RUN pinned below, instead of one long scrolling column. The Write draft is readable; it used to clip every beat at 110 characters.
- lock and a render range on each shot card. Freeze a take you like and re-render only the hops around it.
- Correct resolutions. H3 is a 768-short-edge model, so the top rung at 16:9 is 1344×768 — the size MiniMax states as native. Eleven aspect ratios across five tiers.
- Reference clips take a description, plus a video input size so a 4K plate isn't decoded at full resolution just to be scaled down.
- Changing one reference re-renders only the hops it rides, not the chain.
V1.01
SMALL BUGFIXES, UPDATE TO 1.01 IF YOU HAVEN'T. (mostly to do with the LM integration and some reference file duplication in the input folder)
Hand Tie Clips v1.0 — talking video that keeps going past one clip.
MiniMax H3 gives you one short clip with sound. This node chains several into one continuous take — same face, same room, dialogue carrying across the joins — from a single queue.
▎ New in 1.0
LM Studio integration
You don't have to write the plan. Describe your scene in a sentence, press Write plan, and it fills in the script and your reference settings, reading the photos you've dropped in rather than guessing from filenames. Treat it as a good first draft — worth a read before you queue, and much faster than starting cold.
Soundtrack input
A music bed. Drop in a track and it sits under the whole chain, ducking automatically under the dialogue. Loop or play once, with fades.
Trim bars on every audio and video input. Drag to choose the section you want. This one matters: H3 encodes your whole voice file no matter how long it is, so a three-minute take is a tax on every second of every clip.
The writer model stays warm between plans and is evicted the moment you queue a render, so it doesn't compete with the sampler for VRAM.
Also fixed: no more invented dialogue over the opening seconds of a clip, non-English lines timed correctly, and reference images keeping the size you set.
Search Hand Tie Clips in ComfyUI Manager.