基础功能 基础功能 LLM 调用 支持多种模型接口，比如 OpenAI、Hugging Face、AzureOpenAI … Fake LLM，用于测试 缓存的支持，比如 in-mem（内存）、SQLite、Redis、SQL 用量记录 支持流模式（就是一个字一个字的返回，类似打字效果） Prompt管理，支持各种自定义模板 拥有大量的文档加载...

通过 Google 搜索并返回答案 通过 Google 搜索并返回答案 接下来，我们就来搞点有意思的。我们来让我们的 OpenAI api 联网搜索，并返回答案给我们。 这里我们需要借助 Serpapi 来进行实现，Serpapi 提供了 google 搜索的 api 接口。 首先需要我们到 Serpapi 官网上注册一个用户，https://se...

Register an agent Path and HTTP methods Request fields Example request: Flow agent Example request: Conversational flow agent Example request: Conversational agent Example resp...

Model guide Model guide See the table below for the current recommended models. These models all allow commercial use and offer a blend of speed and performance. Component Mo...

Retrieval-augmented generation processor Request fields Context field list Example Creating a search pipeline Using a search pipeline Retrieval-augmented generation process...

AWS Bedrock LLM Connecting to AWS Bedrock Obtain IAM credentials to use for Bedrock Copy the desired model you wish to use. AWS Bedrock IAM Access ID AWS Bedrock IAM Access Key ...

Use Cases Semantic Search LLM Orchestration Language Model Workflows Use Cases The following sections introduce common txtai use cases. A comprehensive set of over 50 example...

Neural Sparse Search tool Step 1: Register and deploy a sparse encoding model Step 2: Ingest data into an index Step 3: Register a flow agent that will run the NeuralSparseSearch...

Configuring model guardrails Prerequisites Step 1: Create a guardrail index Step 2: Index excluded words or phrases Step 3: Register a model group Step 4: Create a connector S...

Workflow Example Configuration-driven example LLM workflow example Methods More examples Workflow Workflows are a simple yet powerful construct that takes a callable and...