Anura R1 Ultra
Copyright (c) 2026 Genysis IQ

Anura R1 Ultra is built by fine-tuning an open-weight foundation model. This NOTICE
records the third-party works it is derived from or depends on, as their licenses require.

Foundation model
----------------
Qwen3.6-35B-A3B — Copyright (c) Alibaba Cloud (Qwen Team).
Licensed under the Apache License, Version 2.0.
https://huggingface.co/Qwen/Qwen3.6-35B-A3B
MLX 4-bit conversion published by the mlx-community organisation on Hugging Face
(https://huggingface.co/mlx-community/Qwen3.6-35B-A3B-4bit), same license.

Anura's LoRA adapter and its packaged model are modifications of that work: the
foundation weights are unchanged and a low-rank adapter trained by Genysis IQ is applied
at load time.

A copy of the Apache License 2.0 is available at http://www.apache.org/licenses/LICENSE-2.0

Software dependencies (each under its own license)
--------------------------------------------------
MLX, mlx-lm (MIT) — Apple Inc.
mlx-vlm (MIT) — Prince Canuma and contributors
transformers, tokenizers (Apache-2.0) — Hugging Face
FastAPI (MIT), Uvicorn (BSD-3-Clause), httpx (BSD-3-Clause)
trafilatura (Apache-2.0), selectolax (MIT), ddgs (MIT), pypdf (BSD-3-Clause)
Rich (MIT), prompt_toolkit (BSD-3-Clause), Pillow (MIT-CMU), PyYAML (MIT)