NLP Cloud

LiteLLM 支持 NLP Cloud 上的所有 LLM。

API 密钥

import os 

os.environ["NLP_CLOUD_API_KEY"] = "your-api-key"

示例用法

import os
from litellm import completion 

# set env
os.environ["NLP_CLOUD_API_KEY"] = "your-api-key" 

messages = [{"role": "user", "content": "Hey! how's it going?"}]
response = completion(model="dolphin", messages=messages)
print(response)

流式传输

调用 completion 时，只需设置 stream=True。

import os
from litellm import completion 

# set env
os.environ["NLP_CLOUD_API_KEY"] = "your-api-key" 

messages = [{"role": "user", "content": "Hey! how's it going?"}]
response = completion(model="dolphin", messages=messages, stream=True)
for chunk in response:
    print(chunk["choices"][0]["delta"]["content"])  # same as openai format

非 dolphin 模型

默认情况下，LiteLLM 会将 dolphin 和 chatdolphin 映射到 nlp cloud。

如果您尝试使用 nlp cloud 调用任何其他模型（例如 GPT-J、Llama-2 等），只需将其设置为您的自定义 llm 提供商。

import os
from litellm import completion 

# set env - [OPTIONAL] replace with your nlp cloud key
os.environ["NLP_CLOUD_API_KEY"] = "your-api-key" 

messages = [{"role": "user", "content": "Hey! how's it going?"}]

# e.g. to call Llama2 on NLP Cloud
response = completion(model="nlp_cloud/finetuned-llama-2-70b", messages=messages, stream=True)
for chunk in response:
    print(chunk["choices"][0]["delta"]["content"])  # same as openai format