This is a fine-tuned version of LFM2.5-2.6B using the Skyhigh-2203/MiMo-2.5-Pro-Reasoning-Traces-Hard dataset in an effort to improve reasoning and tool-calling for use with my own hex personal assistant agent. After seeing a positive improvement overall in my own benchmark, I decided to upload the finetune to HF. For what it’s worth, the benchmark scores are available here: hexbench
Модальности:
Генерация текста
Области применения:
Диалог / чат
Задача: Генерация текста
Автор: hxii
Теги: gguf, endpoints_compatible, conversational
Лайков: 3 | Загрузок: 32
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.