Xeretron Turbo 35B — NOTICE
=============================

This file must accompany any distribution of the weights, quantizations
(GGUF/FP8/etc.), adapters, or derivatives of "Xeretron Turbo 35B".

The contents of this NOTICE are informational and do not modify the Apache
License 2.0.


1) Xeretron derivative
----------------------

  Derivative name:      Xeretron Turbo 35B
  Operator / producer:  Xeretron (Xeretron AI Server)
  Modification type:    finetune / post-training and operational packaging
                        (Turbo series: architecture, DevOps, agents,
                        document vision; emphasis on throughput / MTP).
  Modification notice:  This work is a Derivative Work of Qwen3.6-35B-A3B
                        (line with Multi-Token Prediction / MTP).
                        Files and weights have been modified relative to
                        the original Work.

  Copyright © 2026 Xeretron. All rights reserved in Xeretron's own
  contributions (finetune, packaging, product documentation, and the
  "Xeretron Turbo 35B" mark), without prejudice to the rights in the
  base Work.


2) Base Work (upstream)
-----------------------

  Base model:    Qwen3.6-35B-A3B (MTP)
  Authors:       Qwen / Alibaba Cloud (per model card and official repo)
  Identifier:    Qwen/Qwen3.6-35B-A3B
  URL:           https://huggingface.co/Qwen/Qwen3.6-35B-A3B
  License:       Apache License, Version 2.0
                 http://www.apache.org/licenses/LICENSE-2.0

  Technical notes from the base Work (public model card):
  - ~35B total parameters · ~3B active per token (MoE)
  - 256 experts · 8 routed + 1 shared
  - Native context 262,144 · extensible ~1M (YaRN)
  - Vision-language · Multi-Token Prediction

  Copyright © respective owners of Qwen3.6-35B-A3B
  (Qwen / Alibaba Cloud and contributors of the original Work).

  If the upstream repository includes its own NOTICE, that content is
  incorporated here by reference. Check:
  https://huggingface.co/Qwen/Qwen3.6-35B-A3B


3) Trademarks
-------------

  "Qwen", "Qwen3.6" and related marks are property of their respective
  owners. Mention is solely attribution of the base Work.

  "Xeretron", "Xeretron Turbo", "CodiiX" and "Xeretron AI Server" are marks
  of Xeretron. This derivative does not imply sponsorship, endorsement, or
  official association with Alibaba Cloud / Qwen.


4) License / packaging
----------------------

  Include with every weights delivery:
    - LICENSE (Apache 2.0) — see LICENSE-Apache-2.0.en.txt
    - this NOTICE
    - any applicable upstream NOTICE
    - notice that the Work was modified (section 1)

  Recommended metadata:
    base_model: Qwen/Qwen3.6-35B-A3B
    license:    apache-2.0
    tags:       fine-tuned, xeretron, moe, turbo, devops, agents, mtp


5) Contact
----------

  Site:  https://xeretron.com/
  Page:  https://xeretron.com/en/modelo-xeretron-turbo-35b.html