Swift-Qwen3.8-Flash-Next
Copyright 2026 UkisAI

UkisAI's contribution (the "Swift Contribution") is licensed under the
Swift Open License v1.0. See LICENSE.

This model is a derivative work of Qwen3.8-Flash-Next
  https://huggingface.co/Qwen/Qwen3.8-Flash-Next
  Copyright (c) 2026 Qwen
  Licensed under the Qwen Community License 1.0. See LICENSE-QWEN.

Changes made by UkisAI:
  - model-*.safetensors: model weights were modified by UkisAI through fine-tuning.
  - tokenizer.json, tokenizer_config.json: re-serialized from UkisAI's training
    base. Vocabulary and merges are unchanged; the file format, some metadata
    fields and the pre-tokenizer pattern differ from the Base Model files.
  - README.md: replaced. LICENSE (Swift Open License v1.0), NOTICE,
    benchmarks/ and ukisai-banner.png added.
  - The Base Model's license text was moved from LICENSE to LICENSE-QWEN.
  - All other files (config.json, generation_config.json, chat_template.jinja,
    model.safetensors.index.json, vocab.json, merges.txt,
    preprocessor_config.json, video_preprocessor_config.json) are unmodified
    from Qwen3.8-Flash-Next and remain under the Qwen Community License 1.0.

Changes made by JRamirez-UAB (this repository, 2026-10-13):
  - model-*.safetensors, model.safetensors.index.json, config.json, quantization_config.json:
    the Swift 1.5 Qwen3.8-Flash-Next BF16 weights were quantized to EXL3 (3.75 bits per weight
    on average) with exllamav3 1.5.1. These files are a derivative of the Swift contribution and
    remain under the Swift Open License v1.0 (and the Qwen Community License 1.0 for the base model).
  - ngram_embedding.safetensors: the n-gram (PLE) table quantized to EXL3.
  - README.md: replaced. scripts/ and plugin/ added (our code, Apache-2.0).
  - Not included (download them from their owners with scripts/assemble.py): NVIDIA's FP8 PLE table
    (nvidia/Qwen3.8-Flash-Next-NVFP4, NVIDIA Open Model License) and turboderp's MTP mixer patch
    (turboderp/Qwen3.8-Flash-Next-exl3, branch 3.05bpw_h5_ng5).