ChatOCIModelDeployment

这将帮助您开始使用 OCIModelDeployment 聊天模型。有关 ChatOCIModelDeployment 功能和配置的详细文档，请参阅 API 参考。

OCI Data Science 是一个完全托管的无服务器平台，供数据科学团队在 Oracle Cloud Infrastructure 中构建、训练和管理机器学习模型。您可以使用 AI Quick Actions 在 OCI Data Science 模型部署服务上轻松部署 LLM。您可以选择使用流行的推理框架（如 vLLM 或 TGI）来部署模型。默认情况下，模型部署端点模仿 OpenAI API 协议。

有关最新更新、示例和实验性功能，请参阅 ADS LangChain 集成。

概述

集成详情

类	包	本地	可序列化	JS 支持	包下载	包最新
ChatOCIModelDeployment	langchain-community	❌	beta	❌

模型功能

工具调用	结构化输出	JSON 模式	图像输入	音频输入	视频输入	令牌级流式传输	原生异步	令牌使用情况	Logprobs
依赖	依赖	依赖	依赖	依赖	依赖	✅	✅	✅	✅

某些模型功能，包括工具调用、结构化输出、JSON 模式和多模态输入，取决于已部署的模型。

设置

要使用 ChatOCIModelDeployment，您需要部署一个具有聊天完成端点的聊天模型，并安装 langchain-community、langchain-openai 和 oracle-ads 集成包。

您可以使用 OCI Data Science 模型部署上的 AI Quick Actions 轻松部署基础模型。有关其他部署示例，请访问 Oracle GitHub 示例存储库。

策略

请确保您拥有访问 OCI Data Science 模型部署端点所需的策略。

凭证

您可以通过 Oracle ADS 设置身份验证。当您在 OCI Data Science Notebook Session 中工作时，可以利用资源主体来访问其他 OCI 资源。

import ads

# Set authentication through ads
# Use resource principal are operating within a
# OCI service that has resource principal based
# authentication configured
ads.set_auth("resource_principal")

或者，您可以使用以下环境变量配置凭据。例如，要使用具有特定配置文件的 API 密钥：

import os

# Set authentication through environment variables
# Use API Key setup when you are working from a local
# workstation or on platform which does not support
# resource principals.
os.environ["OCI_IAM_TYPE"] = "api_key"
os.environ["OCI_CONFIG_PROFILE"] = "default"
os.environ["OCI_CONFIG_LOCATION"] = "~/.oci"

请查看 Oracle ADS 文档 以了解更多选项。

安装

LangChain OCIModelDeployment 集成位于 langchain-community 包中。以下命令将安装 langchain-community 和所需的依赖项。

%pip install -qU langchain-community langchain-openai oracle-ads

实例化

您可以使用通用的 ChatOCIModelDeployment 或特定框架的类，如 ChatOCIModelDeploymentVLLM 来实例化模型。

• 当您需要部署模型的通用入口点时，请使用 ChatOCIModelDeployment。您可以在实例化此类时通过 model_kwargs 传递模型参数。这提供了灵活性和便捷的配置，而无需依赖特定框架的细节。

from langchain_community.chat_models import ChatOCIModelDeployment

# Create an instance of OCI Model Deployment Endpoint
# Replace the endpoint uri with your own
# Using generic class as entry point, you will be able
# to pass model parameters through model_kwargs during
# instantiation.
chat = ChatOCIModelDeployment(
    endpoint="https://modeldeployment.<region>.oci.customer-oci.com/<ocid>/predict",
    streaming=True,
    max_retries=1,
    model_kwargs={
        "temperature": 0.2,
        "max_tokens": 512,
    },  # other model params...
    default_headers={
        "route": "/v1/chat/completions",
        # other request headers ...
    },
)

API Reference:ChatOCIModelDeployment

• 使用特定框架的类，如 ChatOCIModelDeploymentVLLM：当您使用特定框架（例如 vLLM）并需要直接通过构造函数传递模型参数时，这非常适用，可以简化设置过程。

from langchain_community.chat_models import ChatOCIModelDeploymentVLLM

# Create an instance of OCI Model Deployment Endpoint
# Replace the endpoint uri with your own
# Using framework specific class as entry point, you will
# be able to pass model parameters in constructor.
chat = ChatOCIModelDeploymentVLLM(
    endpoint="https://modeldeployment.<region>.oci.customer-oci.com/<md_ocid>/predict",
)

API Reference:ChatOCIModelDeploymentVLLM

调用

messages = [
    (
        "system",
        "You are a helpful assistant that translates English to French. Translate the user sentence.",
    ),
    ("human", "I love programming."),
]

ai_msg = chat.invoke(messages)
ai_msg

AIMessage(content="J'adore programmer.", response_metadata={'token_usage': {'prompt_tokens': 44, 'total_tokens': 52, 'completion_tokens': 8}, 'model_name': 'odsc-llm', 'system_fingerprint': '', 'finish_reason': 'stop'}, id='run-ca145168-efa9-414c-9dd1-21d10766fdd3-0')

print(ai_msg.content)

J'adore programmer.

Chaining

from langchain_core.prompts import ChatPromptTemplate

prompt = ChatPromptTemplate.from_messages(
    [
        (
            "system",
            "You are a helpful assistant that translates {input_language} to {output_language}.",
        ),
        ("human", "{input}"),
    ]
)

chain = prompt | chat
chain.invoke(
    {
        "input_language": "English",
        "output_language": "German",
        "input": "I love programming.",
    }
)

API Reference:ChatPromptTemplate

AIMessage(content='Ich liebe Programmierung.', response_metadata={'token_usage': {'prompt_tokens': 38, 'total_tokens': 48, 'completion_tokens': 10}, 'model_name': 'odsc-llm', 'system_fingerprint': '', 'finish_reason': 'stop'}, id='run-5dd936b0-b97e-490e-9869-2ad3dd524234-0')

异步调用

from langchain_community.chat_models import ChatOCIModelDeployment

system = "You are a helpful translator that translates {input_language} to {output_language}."
human = "{text}"
prompt = ChatPromptTemplate.from_messages([("system", system), ("human", human)])

chat = ChatOCIModelDeployment(
    endpoint="https://modeldeployment.us-ashburn-1.oci.customer-oci.com/<ocid>/predict"
)
chain = prompt | chat

await chain.ainvoke(
    {
        "input_language": "English",
        "output_language": "Chinese",
        "text": "I love programming",
    }
)

API Reference:ChatOCIModelDeployment

AIMessage(content='我喜欢编程', response_metadata={'token_usage': {'prompt_tokens': 37, 'total_tokens': 50, 'completion_tokens': 13}, 'model_name': 'odsc-llm', 'system_fingerprint': '', 'finish_reason': 'stop'}, id='run-a2dc9393-f269-41a4-b908-b1d8a92cf827-0')

流式通话

import os
import sys

from langchain_community.chat_models import ChatOCIModelDeployment
from langchain_core.prompts import ChatPromptTemplate

prompt = ChatPromptTemplate.from_messages(
    [("human", "List out the 5 states in the United State.")]
)

chat = ChatOCIModelDeployment(
    endpoint="https://modeldeployment.us-ashburn-1.oci.customer-oci.com/<ocid>/predict"
)

chain = prompt | chat

for chunk in chain.stream({}):
    sys.stdout.write(chunk.content)
    sys.stdout.flush()

API Reference:ChatOCIModelDeployment | ChatPromptTemplate

1. California
2. Texas
3. Florida
4. New York
5. Illinois

结构化输出

from langchain_community.chat_models import ChatOCIModelDeployment
from pydantic import BaseModel


class Joke(BaseModel):
    """A setup to a joke and the punchline."""

    setup: str
    punchline: str


chat = ChatOCIModelDeployment(
    endpoint="https://modeldeployment.us-ashburn-1.oci.customer-oci.com/<ocid>/predict",
)
structured_llm = chat.with_structured_output(Joke, method="json_mode")
output = structured_llm.invoke(
    "Tell me a joke about cats, respond in JSON with `setup` and `punchline` keys"
)

output.dict()

API Reference:ChatOCIModelDeployment

{'setup': 'Why did the cat get stuck in the tree?',
 'punchline': 'Because it was chasing its tail!'}

API 参考

有关所有功能和配置的详细信息，请参阅每个类的 API 参考文档：

• ChatOCIModelDeployment
• ChatOCIModelDeploymentVLLM
• ChatOCIModelDeploymentTGI

Related

• Chat model conceptual guide
• Chat model how-to guides