RuCoverLAB/deepseek-r1-8b-abliterated-q4-Q4_K_M-GGUF

This model was converted to GGUF format from RuCoverLAB/deepseek-r1-8b-abliterated-q4 using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model.

English-language instructions:

How can it be used?

Install Ollama or Llama.cpp (works on MacOS, Windows and Linux) and download this model! or Download directly via the Hugging Face web interface or using the Hugging Face CLI.

Instructions for Ollama:

CLI:

ollama run RuCoverLAB/deepseek-r1-8b-abliterated-q4

cURL:

curl http://localhost:11434/api/chat \
  -d '{
    "model": "RuCoverLAB/deepseek-r1-8b-abliterated-q4",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Python:

from ollama import chat

response = chat(
    model='RuCoverLAB/deepseek-r1-8b-abliterated-q4',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

JavaScript:

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'RuCoverLAB/deepseek-r1-8b-abliterated-q4',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Instructions for Hugging Face CLI:

CLI:

hf download RuCoverLAB/deepseek-r1-8b-abliterated-q4-Q4_K_M-GGUF --local-dir "%Replace %% with the path to the folder where the model weights will be downloaded.%"

---

P.S. Please subscribe to my Telegram channel! Here is the link: https://t.me/rucoverlab

Русско язычная инструкция:

Как я могу запустить и/или скачать модель?

Установи Ollama или Llama.cpp (работает на MacOS, Windows, Linux) и скачай эту модель! или Скачай модель с веб интерфейса Hugging Face или с помощью Hugging Face CLI.

Инструкции для Ollama

CLI:

ollama run RuCoverLAB/deepseek-r1-8b-abliterated-q4

cURL:

curl http://localhost:11434/api/chat \
  -d '{
    "model": "RuCoverLAB/deepseek-r1-8b-abliterated-q4",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Python:

from ollama import chat

response = chat(
    model='RuCoverLAB/deepseek-r1-8b-abliterated-q4',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

JavaScript:

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'RuCoverLAB/deepseek-r1-8b-abliterated-q4',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Инструкция для Hugging Face CLI:

CLI:

hf download RuCoverLAB/deepseek-r1-8b-abliterated-q4-Q4_K_M-GGUF --local-dir "%Replace %% with the path to the folder where the model weights will be downloaded.%"

---

P.S. Пожалуйста подпишитесь на мой Telegram канал! Вот ссылка: https://t.me/rucoverlab