Loading
Have a ComfyUI workflow? Get a readable Workflow Doc in your browser.Make a workflow readable.Try it now ↗
Image-to-video, a long-form memory sampler, and a VRAM fix that cuts render time roughly 4x on 32 GB cards.
Runtime profile
Source description
v1.1
Image-to-video, a long-form memory sampler, and a VRAM fix that cuts render time roughly 4x on 32 GB cards.
Image-to-video
The sampler now takes an optional start_image . The AIO workflow ships with LoadImage → H3 Optional Image → start_image , so one graph does both: flip the toggle on to begin from your frame, off for pure text-to-video. Shot 1 keeps its first frame, so the image you supply is the image you get.
Why a custom toggle? A normal switch node cannot express “no image” — both of its branches are required — so turning I2V off that way ends up feeding a black placeholder frame. That is not text-to-video, it is video that starts from black. H3 Optional Image emits nothing when disabled.
AI-generated commentary
AI-generated explanation based on source and configuration details. Suggestions are clearly labeled.
This MiniMax H3 workflow turns a story or shot prompts into a multi-shot video with sound, with optional photo and voice inputs.
With the AIO image toggle on, shot one starts from your supplied frame and keeps it as its first frame; with it off, the graph runs text-to-video.
MEMORY mode continues each shot from the latest frame while showing the encoder a fixed opening anchor plus recent shot-end frames.
Add a story or shot prompts, with an optional opening photo and short voice clip; the graph saves a multi-shot video with sound.
Use ComfyUI and the MiniMax-H3 Multishot node pack, available through Manager or the version zip, plus a MiniMax H3 checkpoint.
The AIO and MEMORY variants list minimax_h3_audio_vae_fp32.safetensors, minimax_h3_video_vae_fp16.safetensors, MiniMax-H3-encoder-Q4_K_M.gguf, and MiniMax-H3-fl2va-Q5_1.gguf.
The 3-chain expert variant lists minimax_h3_audio_vae_fp32.safetensors, MiniMax-H3-fl2va-Q5_1.gguf, and qwen3vl_32b_minimax_h3_int8_convrot.safetensors.
Set anchor_frames to 1 to keep the opening anchor active; memory_frames controls how many recent shot-end frames the encoder also sees, and 0 gives stock behavior.
The parser auto-closes scripts with a missing brace or bracket, trailing comma, or unterminated string and shows a warning; other malformed scripts still fail.
Multi-shot needs an mmproj sidecar because chained frames pass through the encoder's vision path.
Does this need to be edited?
Sign in to send an edit request.
Sources
1 sourceSource excerpts
1 excerptSource context: 10753 downloads · Type Workflows · Base model MiniMax H3
~4x faster on 32 GB cards
The text encoder is now evicted before sampling. The Qwen3-VL encoder (~16.5 GB even at Q4) and the H3 DiT (~25 GB) do not co-fit on a 32 GB card, so the DiT was loading partially and streaming ~19 GB from system RAM on every step. If you ever saw this in your log, that was it:
loaded partially; 6423 MB usable, 5847 MB loaded, 19363 MB offloaded
Conditioning is already computed before sampling starts, so the encoder is safe to evict there. Measured on an RTX 5090: ~60 min → ~15 min for the same render. The node prints [H3Multishot] TE evicted; NN.N GB free for the DiT each shot so you can confirm it.
New: long-form memory sampler
For 2–5 minute videos (12–30 shots), where plain chaining drifts. Stock chaining shows each shot one image — the previous shot’s last frame — so every hop can only see one hop back and identity error compounds.
H3 Multishot Sampler + Memory splits the two jobs stock chaining conflates:
Keyframe — what the video physically continues from. Still the most recent frame, so seams stay smooth.
Memory — what the encoder actually looks at : a persistent anchor from the start of the piece plus the last N shot-end frames. The anchor never changes, so drift cannot compound.
Knobs: anchor_frames (1 = on, the long-chain fix) and memory_frames (recent frames the encoder also sees; 0 = stock behaviour). H3’s encoder takes multiple images natively, so this uses the model’s own mechanism, just deeper.
Script parser self-repairs
Long JSON scripts that lose their closing brace or bracket, end on a trailing comma, or leave a string unterminated are now auto-closed with a console warning instead of failing the render. A 4,500-character script missing one } is not a typo the author can see. Genuinely malformed scripts still fail loudly.
Tip from testing
Render at H3’s native resolution and upscale afterwards. Rendering natively at 1920x1088 scored worse than 960x544 in blind review — softer detail, and a reviewer called it “looks upscaled” — while costing about 4x the time.
Models
DiT quants — joeygambino/MiniMax-H3-GGUF
Text encoder + mmproj — 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.
VAEs — Comfy-Org/MiniMax-H3
Support
Everything here is free and stays free. If it saved you a night of debugging (it contains several hundred of mine), tips keep the 5090 warm: Ko-fi · GitHub Sponsors · Liberapay . nsors · Liberapay .
Hardware and download-size estimates are not available yet.
Requirements
minimax_h3_audio_vae_fp32.safetensors
File unverifiedVAE · Hugging Face · Comfy-Org/MiniMax-H3 · MiniMax H3
MiniMax-H3-fl2va-Q5_1.gguf
Not resolvedCheckpoint · Unknown
Text encoder · Hugging Face · Comfy-Org/MiniMax-H3 · MiniMax H3
