Creator's Loop Pro 0.12.0 - Mobile, Private AI, Voice, CPU Safe, and AI Media Third-Party Notices

WebLLM
Copyright the MLC/WebLLM contributors.
Licensed under the Apache License, Version 2.0.
Project: https://github.com/mlc-ai/web-llm
The complete license text is included at ./web-llm-APACHE-2.0.txt.

On-device models
Creator's Loop does not package AI model weights in the application. When a customer explicitly activates Private AI, the selected model is downloaded from the model host and cached by that customer's browser. Model repositories contain their own license and attribution information. SocialFuse should review the selected model license before changing or adding a production model.

Qwen3 WebLLM model repositories:
https://huggingface.co/mlc-ai/Qwen3-0.6B-q4f16_1-MLC
https://huggingface.co/mlc-ai/Qwen3-1.7B-q4f16_1-MLC
https://huggingface.co/mlc-ai/Qwen3-4B-q4f16_1-MLC

Wllama CPU Safe runtime
Copyright (c) 2023 ngxson and contributors.
Licensed under the MIT License.
Project: https://github.com/ngxson/wllama
The complete license text is included at ./wllama-MIT.txt.

Qwen3 0.6B CPU Safe model
Creator's Loop does not package this model weight. After customer approval, Wllama downloads the official Qwen Q8_0 GGUF file and caches it in that customer's browser. GPU layers are explicitly disabled for CPU Safe processing. The model is licensed under Apache-2.0.
https://huggingface.co/Qwen/Qwen3-0.6B-GGUF

Meta Llama 3.2 Model Lab alternatives
Creator's Loop does not package these model weights. A customer may explicitly select and download a compatible quantized model for comparison in Model Lab. Llama models are provided under the Meta Llama 3.2 Community License. Review the repository license and acceptable-use policy before production use.
https://huggingface.co/mlc-ai/Llama-3.2-1B-Instruct-q4f16_1-MLC
https://huggingface.co/mlc-ai/Llama-3.2-3B-Instruct-q4f16_1-MLC
https://www.llama.com/llama3_2/license/

Transformers.js
Copyright the Hugging Face Transformers.js contributors.
Licensed under the Apache License, Version 2.0.
Project: https://github.com/huggingface/transformers.js

Whisper tiny.en on-device speech model
Based on OpenAI Whisper, licensed under the MIT License, and converted to ONNX for Transformers.js.
Model: https://huggingface.co/onnx-community/whisper-tiny.en
Creator's Loop does not package the speech model weights. They download only after the customer approves private voice input and are cached by that customer's browser. Recorded audio is processed on the device and discarded after transcription.