TinyServe, a demo by Tenzan Logic. Notices for the language models it loads in the visitor's browser.
The weights are downloaded by the visitor's browser from huggingface.co; this site does not host them.

Built with Llama.
Llama 3.2 is licensed under the Llama 3.2 Community License, Copyright © Meta Platforms, Inc. All Rights Reserved.
License text: llama-3.2-community-license.txt (this folder)
Acceptable Use Policy: https://www.llama.com/llama3_2/use-policy
Model: onnx-community/Llama-3.2-1B-Instruct (ONNX conversion of meta-llama/Llama-3.2-1B-Instruct).

LFM2.5 1.2B Thinking is licensed under the LFM Open License v1.0, by Liquid AI, Inc.
License text: lfm-open-license-1.0.txt (this folder)
Model: LiquidAI/LFM2.5-1.2B-Thinking-ONNX.
Tenzan Logic is far below the license's threshold of 10 000 000 USD annual revenue (section 1, "Threshold"; section 5).

Qwen3 and DeepSeek-R1 distills: Apache-2.0 and MIT, as the model cards on huggingface.co state.
This page does not state or imply that Meta, Liquid AI or any other model author endorses TinyServe.