Google Veo3.1 Lite Start End To Video (Official Edition)

The display name says it all again: "Google Veo3.1 Lite Start End To Video (Official Edition)." It's the first-last-frame variant on the budget tier - start image, end image, video between - with the full Official Edition controls, priced for the kind of iterating that would break the bank on Pro.

This is the node for a very specific, very common job: you've got a storyboard with two key frames and you need to see how the transition plays out, over and over, without the cost compounding. Lite FLF gives you duration, generate_audio, negative_prompt, seed, and aspect_ratio on top of the frame pair, so the iterations you run are real experiments, not blind dice rolls.

Why you'd reach for it

Because FLF is the most controllable Veo mode, and Lite is the tier where control doesn't cost like a flagship. Lock a seed to hold a transition steady while you tweak the prompt; throw a negative_prompt at the warp or flicker you keep seeing; keep duration short while you're testing. When the transition finally works on Lite, that's when you spend up for the final pass on a higher tier.

How it works

Required: first_frame_url (IMAGE - the opening frame), prompt, resolution (720p default), duration (default 4), generate_audio (default true). Optional: last_frame_url (IMAGE - the destination frame), negative_prompt, aspect_ratio (16:9 default), seed (0 default). The inputs are named for the provider's URL-based API contract, but you wire in ordinary IMAGE tensors and the node handles the upload side.

On queue the node submits both frames to Veo 3.1 Lite through BizyAir's channel, polls, and returns video (VIDEO - preview or save like any local render), request_id (STRING), and urls (STRING). Optional api_config (BIZYAIR_OPENAPI_CONFIG) overrides base_url/api_key per request; skip_error turns a failed call into a placeholder instead of an aborted queue.

Install and setup

One-time install, identical across the pack:

cd ComfyUI/custom_nodes
git clone https://github.com/siliconflow/BizyAirPlus.git
cd BizyAirPlus
python -m pip install -r requirements.txt

Restart ComfyUI (or install "BizyAirPlus" via ComfyUI Manager). No model files - the bizyair-cloudberry and bizytrd pip packages auto-install on startup. Click the BizyAirPlus button to ON, get an API key from bizyair.ai, enter it, and queue.

Common issues

• Soft transitions - Lite is the budget tier; use it to validate the idea, not to judge final quality. A promising-but-soft result means "re-render on a higher tier," not "debug."
• Frames disagree - FLF bridges plausibly, not deterministically; mismatched lighting between start and end frames makes the model invent a jarring middle. Match the frames first.
• Unwanted audio - generate_audio defaults on; flip it off for silent test clips.

Pack caveat: BizyAir auto-installs and sends your images to the cloud. Legitimate SiliconFlow product, but keep the switch OFF for local-only sessions.