AI & Model Disclaimer
Local generation gives you control, not certainty. Review every output and every model license.
AI output
Generated video and audio may contain errors, artifacts, stereotypes, unsafe material, unintended resemblance, fabricated details, or content inconsistent with a prompt. Output is not professional, medical, legal, financial, safety, or factual advice. Human review is required before use or publication.
Model support
Catalog support means the app has an adapter for the declared architecture and pipeline; it is not a guarantee that every checkpoint, quantization, custom fork, LoRA, or future repository revision will work. Performance depends on unified memory, available memory, disk capacity, thermals, macOS, model size, dimensions, frame count, and third-party runtime behavior.
Custom Hugging Face support is intentionally adapter-based. Adding a repository does not make an unknown architecture compatible and does not execute remote custom Python code.
Licenses and rights
BytesWeavers does not own or license third-party checkpoints for you. A model’s presence in the catalog does not authorize commercial use. Review the repository’s current model card, license, acceptable-use policy, training-data restrictions, attribution requirements, and territorial terms. You are responsible for rights to source images, video, audio, likenesses, trademarks, and output.
Named projects
MiniMax, LTX, Wan, Hugging Face, Diffusers, MLX, Apple, Mac, and other names are trademarks or project names of their respective owners. BytesWeavers is not affiliated with or endorsed by those third parties unless expressly stated.
Questions
See the Terms of Use, Privacy Policy, or contact bytesweavers@gmail.com.