Petals
Petals: https://github.com/bigscience-workshop/petals
先决条件
确保已安装 petals
pip install git+https://github.com/bigscience-workshop/petals
用法
确保为所有 petals LLM 添加 petals/
作为前缀。这将把 custom_llm_provider 设置为 petals。
from litellm import completion
response = completion(
model="petals/petals-team/StableBeluga2",
messages=[{ "content": "Hello, how are you?","role": "user"}]
)
print(response)
流式传输用法
response = completion(
model="petals/petals-team/StableBeluga2",
messages=[{ "content": "Hello, how are you?","role": "user"}],
stream=True
)
print(response)
for chunk in response:
print(chunk)
模型详情
模型名称 | 函数调用 |
---|---|
petals-team/StableBeluga | completion('petals/petals-team/StableBeluga2', messages) |
huggyllama/llama-65b | completion('petals/huggyllama/llama-65b', messages) |