slybase/Swift-1.5-Qwen3.8-27b-heretic-MXFP4-DFlash2-radiance - Каталог нейросетей
Генерация текста

slybase/Swift-1.5-Qwen3.8-27b-heretic-MXFP4-DFlash2-radiance

Добавлено:
slybase/Swift-1.5-Qwen3.8-27b-heretic-MXFP4-DFlash2-radiance

> [!CAUTION] > Research artifact. Safety alignment has been deliberately removed in the source model > (akumaburn/Swift-1.5-Qwen3.8-27b-heretic). > This quantised build inherits that: it will attempt to comply with harmful requests the original > model refuses. Provided for research only, with no warranty and no liability. > [!IMPORTANT] > Requires Radiance 1.2.3 or newer. This container was converted with the Radiance 1.2.3 recipe > and does not load on older builds. Get Radiance from > StillDeadcode/radiance (tag v1.2.3 or later). A single Radiance container (.rad, 18.05 GiB) holding the target model and its speculative drafter, quantised for AMD RDNA4 (Radeon AI PRO R9700, gfx1201, 32 GB). It is not a Transformers/vLLM checkpoint: it runs only with the Radiance inference engine. The GPTQ source checkpoint is published separately as slybase/Swift-1.5-Qwen3.8-27B-heretic-MXFP4-GPTQ. 1. Qwen/Qwen3.8-27B (Apache-2.0) 2. ukisai/Swift-1.5-Qwen3.8-27b (Swift Open License v1.0) 3. akumaburn/Swift-1.5-Qwen3.8-27b-heretic (abliterated, BF16) 4. GPTQ to MXFP4 (llm-compressor, MXFP4A16, static act-order, 89 calibration samples, lmhead/vision/MTP excluded — gptq-recipe.yaml) 5. rad-convert…

Модальности:
Генерация текста

Языки программирования:
Swift


Задача: Генерация текста
Автор: slybase
Теги: radiance, mxfp4, gptq, dflash2, speculative-decoding, qwen3.8, swift-1.5, heretic
Лайков: 3  |  Загрузок: 0

Открыть на HuggingFace →

Описание основано на материалах HuggingFace. Перевод выполнен автоматически.