Concurrent Submit | zhenzhen-happyhorse-1.1-lowprice

HappyHorse 1.1 is a newer ByteDance-family video model, and this is the node that hands you all three entry points - text-to-video, image-to-video, and reference-to-video - in one dropdown, on the domestic low-price storefront, batched ten at a time. For a model this new, the pack's pattern of "cheap drafts first" is the right way to learn what it's actually good at.

What's genuinely useful here is the r2v (reference-to-video) mode. Feed up to nine reference images and HappyHorse will try to hold that style/character across the generated motion - the pack added these nodes alongside its seedance-style multi-reference work, so the consistency story is the selling point. If you've been character-locked by your usual model, this is a cheap way to test a different one.

How the concurrent wrapper works

The plain Comfly_happyhorse_1_1_lowprice node is a Seedance2 Low Price unified node: it takes an api_config (ZHENZHEN_SEEDANCE2_CONFIG from the pack's settings node) for the domestic URL and key, assembles references, submits, polls, and downloads the clip. The Submit wrapper keeps every input but returns a single task of type COMFLY_VIDEO_FUTURE, deferring the call to the video worker pool (default 10, COMFLY_VIDEO_CONCURRENCY env var).

The task output flows into ComflyConcurrent_Video_Await, which waits for all connected tasks and returns video_1 … video_10 in order plus a status JSON. failure_mode (fail_fast / placeholder) decides whether one failure kills the batch or just leaves a blank slot.

Inputs that matter

model is required: happyhorse-1.1-t2v, happyhorse-1.1-i2v, or happyhorse-1.1-r2v. prompt is required. seconds runs 3–15 (default 4 - short clips are the cheap way to probe the model). resolution is 720p or 1080p; ratio is adaptive by default with the usual six plus 21:9.

image1–image9 are the reference inputs - all of them matter for r2v and i2v, none for t2v. api_config connects the pack's domestic settings node (that's how the low-price family authenticates; there's no bare api_key field here). seed is cache-control only. skip_error (false by default) returns an empty result instead of raising if a call fails.

Install and gotchas

Pack install as always: ComfyUI Manager → search Comfyui-zhenzhen, or git clone https://github.com/T8mars/Comfyui-zhenzhen into ComfyUI/custom_nodes, restart. No local models - the weights live on Zhenzhen's servers.

The low-price gotcha applies: this node needs the domestic api.seedance.nz key from its own settings node, not your overseas key. Also watch the mode/reference pairing - feeding nine images to t2v is harmless but pointless, and feeding none to r2v will just error. And because it's a fresh model, expect occasional upstream 500s while the backend settles; the pack's standing advice for those is to re-run.