跳到主要内容

LlamaIndex

LlamaIndex 是使用智能体和工作流在您的数据上构建 LLM 驱动智能体的领先框架。 官方网站：llamaindex.ai

1.1 使用方法​

方式一：OpenAILike（推荐）​

from llama_index.llms.openai_like import OpenAILike

llm = OpenAILike(
    model="minimax/minimax-m2.5",
    api_base="https://api.luchentech.com/inference/v1",
    api_key="your-luchentech-key",
    is_chat_model=True
)

response = llm.complete("Hello, how are you?")
print(response.text)

方式二：OpenAI 类​

from llama_index.llms.openai import OpenAI

llm = OpenAI(
    model="minimax/minimax-m2.5",
    api_base="https://api.luchentech.com/inference/v1",
    api_key="your-luchentech-key"
)

方式三：全局默认 LLM​

from llama_index import Settings
from llama_index.llms.openai_like import OpenAILike

Settings.llm = OpenAILike(
    model="minimax/minimax-m2.5",
    api_base="https://api.luchentech.com/inference/v1",
    api_key="your-luchentech-key",
    is_chat_model=True,
    is_function_calling_model=True
)

方式四：RAG 流水线​

from llama_index import VectorStoreIndex, SimpleDirectoryReader
from llama_index.llms.openai_like import OpenAILike

documents = SimpleDirectoryReader("data").load_data()

llm = OpenAILike(
    model="minimax/minimax-m2.5",
    api_base="https://api.luchentech.com/inference/v1",
    api_key="your-luchentech-key",
    is_chat_model=True
)

index = VectorStoreIndex.from_documents(documents, llm=llm)
query_engine = index.as_query_engine()
response = query_engine.query("What is the document about?")

1.2 故障排除​

常见问题解决方案
is_chat_model 错误根据模型类型设置为 True（Chat）或 False（Completion）
函数调用不工作设置 is_function_calling_model=True

1.3 参考​