Support
Everything here is free and stays free — the format spec, the nodes, the workflows, the cartridges, the LoRAs. If it saved you a night of debugging (it contains several hundred of mine), tips keep the 5090 warm:
🔁 Liberapay (recurring)
⚡ Or right here: the Civitai tip button on this page sends Buzz directly.
Type a story. Get one continuous video, with sound. Multi-shot scenes render as a single take - no last-frame chaining, no quality loss from shot to shot. That's the whole pitch.
Both workflows now read left to right: numbered lanes, and you only ever touch lanes 2-4. Everything below the main row is optional.
What you need
ComfyUI + this node pack (Manager: MiniMax-H3 Multishot, or the zip on this version).
A MiniMax-H3 checkpoint (links on this page). 24 GB card? Take a GGUF.
SEAMLESS CHAIN - multi-shot scenes as one take
Lane by lane:
README - the quick start lives on the canvas itself.
1 - MODELS - pick your H3 checkpoint and text encoder. VAEs are preset; LoRA slots are empty until you fill one.
2 - ANCHORS (optional) - a photo to open shot 1 on (enable its gate), and a short voice clip to lock the speaker's voice.
3 - YOUR PROMPTS - type your idea in the box, or point the switch at a prompt file. The writer expands it into shot prompts. Writing your own? Set the writer to
passthrough (raw JSON, skip LLM)and paste shots separated by---lines.4 - CONTROLS - size, frames per shot, steps, and
take_seconds(total length; 30 is a good first run). The switches stay off unless you installed the pack a switch names.5 - ENGINE - nothing to change. The remote encoder lives here if you want the text encoder on a second PC: enter its address, flip the encoder switch, free ~15 GB.
6 - OUTPUT - your video and its audio save here.
Optional panels below the main row: reference images (your character, ref2va checkpoints - folder per character + AUTO REFS on), V2V reference (a clip whose look guides the render), FFLF plates (flf_chain mode only), audio spine (a soundtrack the take follows).
EXTEND TAKE - one person talking, as long as you want
Same lanes, different job: one premise becomes ONE continuous speech cut across windows.
1 - MODELS - same as above.
2 - ANCHORS - a photo of your speaker (shot 1 opens on them) and a voice clip. More useful here than anywhere: one person carries the whole take.
3 - YOUR PROMPT - ONE premise, one speaker. The writer writes the whole speech. num_shots 0 = it decides. Passthrough works here too.
4 - CONTROLS -
take_secondsis the star: 30 ships, 60 clears TikTok's minute.windowstays on auto - it sizes itself to your card.5 - ENGINE / 6 - OUTPUT - same as above.
Keep takes to about 4 windows for now - very long takes slowly sharpen.
Rules of thumb (both workflows)
Spoken lines: 8-12 words per shot. Short lines sync; long lines garble.
Say the sounds you want ("rain on the roof, a fridge hum") or it invents its own.
Keep your character's face in frame - faces carry identity between shots.
If something breaks
Red node? Update the pack in Manager, restart, reload the workflow from disk.
Render crawls at low wattage? Lower resolution or frames per shot, or use the remote encoder.
Only one of the two workflows shows in your sidebar? Fixed in 2.6.5 - re-download both.
Still stuck: comment with your console log. I answer.
Deep dives: the two articles linked on this page. Every lane also has a short note on the canvas.
Detailed guide for people that can read good:
Every setting explained: the Seamless Chain deep manual | Civitai
Description
2.7.0 — per-subject voices, flf_chain fixed, chain leveller, ComfyUI 0.34. Update by replacing both node folders with the ones in this zip; saved workflows load unchanged (every new widget is appended last).
Per-subject voice refs.
voice_ref_2/voice_ref_3on both samplers. Each character keeps their own voice across a chained scene instead of blending into one narrator — verified blind across a full chain, no cross-speaker bleed. This was the most-asked request on this page.flf_chain fix — boundary plates no longer haunted by PLATE0. With
continuity=flf_chainthe memory bank still ran, and its default (bank_pinned=1) pinned a reference clip of shot 1 into every later shot. Shot 1 opens on plate 0, and in flf mode those bank clips are not named in the prompt — so the model treated shot-1 footage as content and mixed PLATE0 back in from shot 2 onward. The bank now stands down automatically in flf_chain and prints a console line saying so. If you hit this on 2.6.x: setbank_pinnedto 0 as a workaround, or update.H3ChainNormalize. New post-chain node: levels the slow texture/colour ratchet across long chains in one pass. Run it after the chain; the known 2.6.0 long-take limit is addressed.
refresh_pin splice alignment. The re-encoded tail was landing 1–2 latent frames early, so joins read as cuts and audio slipped. The splice is now correlation-aligned.
x0 texture clamp dial.
x0_clamp_windowexposed (appended last), dose capped at 0.30. The default is the eye-calibrated setting.Memory-sampler levers. In-loop latent x2 / x1.5 upscale,
refresh_renoise(variance-matched splice),pin_noise_ramp(graded seam floor),auto_chunk_ffn(picks FFN chunking automatically when VRAM is tight).ComfyUI 0.34 supported. 0.34 places interior keyframe anchors natively; the pack probes the running core and stands its own layout patch down when it isn't needed. Older cores keep the patch exactly as before.
Engine-aware writer. The bundled prompt writer now carries separate H3 and LTX system prompts for every mode, selected by a new
enginewidget (appended last). H3 prompts use the native<d>[English] ...</d>spoken-line tag; LTX prompts keep straight-quoted dialogue for the TTS extractor.Small stuff:
beat_secondsrange extended to 20 s on the take controls; a batch of small correctness fixes across both samplers.
Zip contents: both node packs, three workflows (Seamless Chain, Extend Take, Keyframes), INSTALL / SETTINGS / PROMPTING guides, example script. Code first as always: GitHub tag v2.7.0, mirrored on HuggingFace.