Instructions to use cloudyu/60B_MoE_Coder_v3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use cloudyu/60B_MoE_Coder_v3 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="cloudyu/60B_MoE_Coder_v3")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("cloudyu/60B_MoE_Coder_v3")
model = AutoModelForCausalLM.from_pretrained("cloudyu/60B_MoE_Coder_v3", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use cloudyu/60B_MoE_Coder_v3 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "cloudyu/60B_MoE_Coder_v3"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "cloudyu/60B_MoE_Coder_v3",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/cloudyu/60B_MoE_Coder_v3

SGLang

How to use cloudyu/60B_MoE_Coder_v3 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "cloudyu/60B_MoE_Coder_v3" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "cloudyu/60B_MoE_Coder_v3",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "cloudyu/60B_MoE_Coder_v3" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "cloudyu/60B_MoE_Coder_v3",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use cloudyu/60B_MoE_Coder_v3 with Docker Model Runner:
```
docker model run hf.co/cloudyu/60B_MoE_Coder_v3
```

60B_MoE_Coder_v3 / README.md

cloudyu

Update README.md

ccd1289 verified over 2 years ago

preview code

Raw

History Blame Contribute Delete

2.27 kB

	---
	license: other
	tags:
	- yi
	- moe
	license_name: yi-license
	license_link: https://huggingface.co/01-ai/Yi-34B-200K/blob/main/LICENSE
	---
	This is an English and Chinese chat model for Coding but also general purpose.

	example output


	写一个python程序抓取youtube最流行的内容：从网页获取数据，

	以下是一个python脚本，可以从Youtube上抓取最受欢迎的视频并将其保存到本地文件中。该脚本使用Selenium WebDriver来模拟浏览器操作以登录并搜索视频，然后利用Beautiful Soup库提取所需信息。如果您需要任何帮助或修改建议，请随时告诉我！

	```
	from selenium import webdriver
	from bs4 import BeautifulSoup as BS
	import os

	# 设置ChromeDriver路径
	CHROMEDRIVER_PATH = './chromedriver.exe' # 更换为实际路径

	def main():
	# 启动Chrome浏览器并登录Youtube
	browser = webdriver.Chrome(executable_path=CHROMEDRIVER_PATH)
	browser.get('https://www.youtube.com/')
	login_url = 'https://accounts.google.com/signin'
	browser.get(login_url)
	# 模拟输入账号和密码并登录
	username = input('Username: ')
	password = getpass.getpass() # 防止密码显示在终端上
	submit_button = browser.find_element_by_class_name('whsLW', timeout=5)
	submit_button.click()

	# 在搜索栏中输入关键词并查找结果
	search_query = 'most popular youtube videos today'
	search_box = browser.find_element_by_id('search_query')
	search_box.send_keys(search_query)
	enter_key = browser.find_element_by_class_name('ytp-suggest-renderer__input-icon')
	enter_key.click()
	results = BS(browser.page_source, 'html.parser')

	# 从搜索结果中提取出最受欢迎的视频链接并保存到本地文件
	videos = results.find('ol', {'class': 'r'})
	video_urls = [a['href'] for a in videos.find_all('a')]
	saved_path = 'popular_videos/' + os.getcwd().split('/')[-2] + '_' + str(int(time.time())) + '/'
	if not os.path.exists(saved_path):
	os.makedirs(saved_path)
	for i, url in enumerate(video_urls[:50]):
	filename = saved_path + '%d.mp4' % (i+1)
	urllib.request.urlretrieve(url, filename)
	print('Downloaded top 50 videos to:', saved_path)

	except Exception as e:
	print(e)

	```