heathershaw/DeepSeek-R1-Distill-Llama-8B-rk3588-1.1.2 - Каталог нейросетей
Генерация текста

heathershaw/DeepSeek-R1-Distill-Llama-8B-rk3588-1.1.2

Добавлено:
heathershaw/DeepSeek-R1-Distill-Llama-8B-rk3588-1.1.2

This version of DeepSeek-R1-Distill-Llama-8B has been converted to run on the RK3588 NPU using w8a8 quantization. Pretty much anything by these folks: marty1885 and happyme531 Converted using https://github.com/heathershaw821/ez-er-rkllm-toolkit2 We introduce our first-generation reasoning models, DeepSeek-R1-Zero and DeepSeek-R1. DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning (RL) without supervised fine-tuning (SFT) as a preliminary step, demonstrated remarkable performance on reasoning. With RL, DeepSeek-R1-Zero naturally emerged with numerous powerful and interesting reasoning behaviors. However, DeepSeek-R1-Zero encounters challenges such as endless repetition, poor readability, and language mixing. To address these issues and further enhance reasoning performance, we introduce DeepSeek-R1, which incorporates cold-start data before RL. DeepSeek-R1 achieves performance comparable to OpenAI-o1 across math, code, and reasoning tasks. To support the research community, we have open-sourced DeepSeek-R1-Zero, DeepSeek-R1, and six dense models distilled from DeepSeek-R1 based on Llama and Qwen. DeepSeek-R1-Distill-Qwen-32B outperforms OpenAI-o1-mini across…

Модальности:
Генерация текста

Области применения:
Диалог / чат Логика и рассуждение


Задача: Генерация текста
Автор: heathershaw
Теги: llama, conversational, text-generation-inference, endpoints_compatible
Лайков: 4  |  Загрузок: 5

Открыть на HuggingFace →

Описание основано на материалах HuggingFace. Перевод выполнен автоматически.