Dedicated deployment cache of upstream public assets. Contains four hash-pinned base assets from Comfy-Org/MiniMax-H3 plus three pinned DMAD/LightX2V adapters. All weights use the provider native cache; no weights are baked into the worker image. Upstream licenses and use restrictions apply. The standard MiniMax-H3 license excludes the US, EU, UK and South Korea; this mirror does not grant additional rights to recipients. Read LICENSE and NOTICE before use. The text encoder incorporates Apache-2.0-licensed Qwen3-VL-32B: https://github.com/QwenLM/Qwen3-VL/blob/main/LICENSE. Do not update main after qualification; create a new cache repository for changes.
Independent publisher
Cyrus
cyrusaiserver
Models in Library1
Datasets in Library0
Models on Hugging Face1
Followers—
Compare the published evidence
This profile connects the 1 model and dataset record shown below to their upstream cards and sources. The library counts describe SAVRN's coverage, rather than this publisher's entire output.
Reported licenses: other. Open a record to review its specific license and conditions.
Compare model specifications, source dates, and hardware estimates on the individual records. Publisher identity verification is separate from a model's quality, deployment readiness, or suitability for your work.