eino-component

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Eino Component Guide

Eino组件指南

Component Selection Guide

组件选择指南

ChatModel -- LLM inference

ChatModel -- LLM推理

Provider	Package	Notes
OpenAI	`model/openai`	Also supports Azure via `ByAzure: true`
Claude	`model/claude`	Also supports AWS Bedrock via `ByBedrock: true`
Gemini	`model/gemini`	Requires `genai.Client`
Ark (Volcengine)	`model/ark`	Doubao models
Ollama	`model/ollama`	Local models
DeepSeek	`model/deepseek`	Reasoning support
Qwen	`model/qwen`	Alibaba DashScope API
Qianfan	`model/qianfan`	Baidu ERNIE models
OpenRouter	`model/openrouter`	Multi-provider routing

提供商	包名	说明
OpenAI	`model/openai`	也支持Azure，只需设置 `ByAzure: true`
Claude	`model/claude`	也支持AWS Bedrock，只需设置 `ByBedrock: true`
Gemini	`model/gemini`	需要 `genai.Client`
方舟（火山引擎）	`model/ark`	豆包模型
Ollama	`model/ollama`	本地模型
DeepSeek	`model/deepseek`	支持推理
通义千问	`model/qwen`	阿里云DashScope API
千帆	`model/qianfan`	百度ERNIE模型
OpenRouter	`model/openrouter`	多提供商路由

Embedding -- text to vector

Embedding -- 文本转向量

Provider	Package	Notes
OpenAI	`embedding/openai`	text-embedding-3-small/large, ada-002
Ark	`embedding/ark`	Volcengine embedding models
Gemini	`embedding/gemini`	Google embedding models
DashScope	`embedding/dashscope`	Alibaba embedding
Ollama	`embedding/ollama`	Local embedding models
Qianfan	`embedding/qianfan`	Baidu embedding

提供商	包名	说明
OpenAI	`embedding/openai`	支持text-embedding-3-small/large、ada-002模型
方舟	`embedding/ark`	火山引擎Embedding模型
Gemini	`embedding/gemini`	Google Embedding模型
DashScope	`embedding/dashscope`	阿里云Embedding
Ollama	`embedding/ollama`	本地Embedding模型
千帆	`embedding/qianfan`	百度Embedding

Retriever -- vector/keyword search

Retriever -- 向量/关键词搜索

Backend	Package	Notes
Redis	`retriever/redis`	KNN and range vector search
Milvus 2.x	`retriever/milvus2`	Dense + sparse hybrid, BM25
Elasticsearch 8	`retriever/es8`	Approximate vector search
Qdrant	`retriever/qdrant`	Vector similarity search

后端	包名	说明
Redis	`retriever/redis`	支持KNN和范围向量搜索
Milvus 2.x	`retriever/milvus2`	稠密+稀疏混合检索、BM25
Elasticsearch 8	`retriever/es8`	近似向量搜索
Qdrant	`retriever/qdrant`	向量相似度搜索

Indexer -- store documents with vectors

Indexer -- 存储带向量的文档

Backend	Package
Redis	`indexer/redis`
Milvus 2.x	`indexer/milvus2`
Elasticsearch 8	`indexer/es8`
Qdrant	`indexer/qdrant`

后端	包名
Redis	`indexer/redis`
Milvus 2.x	`indexer/milvus2`
Elasticsearch 8	`indexer/es8`
Qdrant	`indexer/qdrant`

Tools -- model-callable functions

Tools -- 模型可调用函数

Tool	Package	Notes
MCP	`tool/mcp`	Model Context Protocol tools
Google Search	`tool/googlesearch`	Custom Search JSON API
DuckDuckGo	`tool/duckduckgo`	Web search (use v2)
Bing Search	`tool/bingsearch`	Bing Web Search API
HTTP Request	`tool/httprequest`	Generic HTTP calls
Command Line	`tool/commandline`	Shell command execution
Browser Use	`tool/browseruse`	Browser automation

工具	包名	说明
MCP	`tool/mcp`	Model Context Protocol工具
Google搜索	`tool/googlesearch`	自定义搜索JSON API
DuckDuckGo	`tool/duckduckgo`	网页搜索（请使用v2版本）
必应搜索	`tool/bingsearch`	必应网页搜索API
HTTP请求	`tool/httprequest`	通用HTTP调用
命令行	`tool/commandline`	Shell命令执行
浏览器操作	`tool/browseruse`	浏览器自动化

Interface Quick Reference

接口快速参考

// ChatModel
type BaseChatModel interface {
    Generate(ctx context.Context, input []*schema.Message, opts ...Option) (*schema.Message, error)
    Stream(ctx context.Context, input []*schema.Message, opts ...Option) (*schema.StreamReader[*schema.Message], error)
}
type ToolCallingChatModel interface {
    BaseChatModel
    WithTools(tools []*schema.ToolInfo) (ToolCallingChatModel, error)
}

// Embedding
type Embedder interface {
    EmbedStrings(ctx context.Context, texts []string, opts ...Option) ([][]float64, error)
}

// Retriever
type Retriever interface {
    Retrieve(ctx context.Context, query string, opts ...Option) ([]*schema.Document, error)
}

// Indexer
type Indexer interface {
    Store(ctx context.Context, docs []*schema.Document, opts ...Option) (ids []string, err error)
}

// Document
type Loader interface {
    Load(ctx context.Context, src Source, opts ...LoaderOption) ([]*schema.Document, error)
}
type Transformer interface {
    Transform(ctx context.Context, src []*schema.Document, opts ...TransformerOption) ([]*schema.Document, error)
}

// Tool
type InvokableTool interface {
    Info(ctx context.Context) (*schema.ToolInfo, error)
    InvokableRun(ctx context.Context, argumentsInJSON string, opts ...Option) (string, error)
}

// Prompt
type ChatTemplate interface {
    Format(ctx context.Context, vs map[string]any, opts ...Option) ([]*schema.Message, error)
}

// ChatModel
type BaseChatModel interface {
    Generate(ctx context.Context, input []*schema.Message, opts ...Option) (*schema.Message, error)
    Stream(ctx context.Context, input []*schema.Message, opts ...Option) (*schema.StreamReader[*schema.Message], error)
}
type ToolCallingChatModel interface {
    BaseChatModel
    WithTools(tools []*schema.ToolInfo) (ToolCallingChatModel, error)
}

// Embedding
type Embedder interface {
    EmbedStrings(ctx context.Context, texts []string, opts ...Option) ([][]float64, error)
}

// Retriever
type Retriever interface {
    Retrieve(ctx context.Context, query string, opts ...Option) ([]*schema.Document, error)
}

// Indexer
type Indexer interface {
    Store(ctx context.Context, docs []*schema.Document, opts ...Option) (ids []string, err error)
}

// Document
type Loader interface {
    Load(ctx context.Context, src Source, opts ...LoaderOption) ([]*schema.Document, error)
}
type Transformer interface {
    Transform(ctx context.Context, src []*schema.Document, opts ...TransformerOption) ([]*schema.Document, error)
}

// Tool
type InvokableTool interface {
    Info(ctx context.Context) (*schema.ToolInfo, error)
    InvokableRun(ctx context.Context, argumentsInJSON string, opts ...Option) (string, error)
}

// Prompt
type ChatTemplate interface {
    Format(ctx context.Context, vs map[string]any, opts ...Option) ([]*schema.Message, error)
}

Installation

安装

bash

go get github.com/cloudwego/eino-ext/components/{type}/{impl}@latest

bash

go get github.com/cloudwego/eino-ext/components/{type}/{impl}@latest

Examples:

go get github.com/cloudwego/eino-ext/components/model/openai@latest go get github.com/cloudwego/eino-ext/components/retriever/milvus2@latest go get github.com/cloudwego/eino-ext/components/tool/mcp@latest

undefined

undefined

ChatModel Usage

ChatModel使用

Generate

生成

resp, err := chatModel.Generate(ctx, []*schema.Message{
    {Role: schema.User, Content: "Hello"},
})
fmt.Println(resp.Content)

resp, err := chatModel.Generate(ctx, []*schema.Message{
    {Role: schema.User, Content: "Hello"},
})
fmt.Println(resp.Content)

Stream

流式输出

reader, err := chatModel.Stream(ctx, messages)
defer reader.Close()
for {
    chunk, err := reader.Recv()
    if errors.Is(err, io.EOF) { break }
    if err != nil { return err }
    fmt.Print(chunk.Content)
}

reader, err := chatModel.Stream(ctx, messages)
defer reader.Close()
for {
    chunk, err := reader.Recv()
    if errors.Is(err, io.EOF) { break }
    if err != nil { return err }
    fmt.Print(chunk.Content)
}

Tool Calling

工具调用

withTools, err := chatModel.WithTools([]*schema.ToolInfo{toolInfo})
resp, err := withTools.Generate(ctx, messages)
// resp.ToolCalls contains model's tool invocations

withTools, err := chatModel.WithTools([]*schema.ToolInfo{toolInfo})
resp, err := withTools.Generate(ctx, messages)
// resp.ToolCalls contains model's tool invocations

RAG Components

RAG组件

Embedding + Indexer + Retriever form the RAG pipeline:

// 1. Embed and store documents
indexer, _ := redisIndexer.NewIndexer(ctx, &redisIndexer.IndexerConfig{
    Client: redisClient, KeyPrefix: "doc:", Embedding: embedder,
})
ids, _ := indexer.Store(ctx, docs)

// 2. Retrieve relevant documents
retriever, _ := redisRetriever.NewRetriever(ctx, &redisRetriever.RetrieverConfig{
    Client: redisClient, Index: "my_index", Embedding: embedder,
})
docs, _ := retriever.Retrieve(ctx, "user query", retriever.WithTopK(5))

Embedding + Indexer + Retriever共同构成RAG流水线：

// 1. 嵌入并存储文档
indexer, _ := redisIndexer.NewIndexer(ctx, &redisIndexer.IndexerConfig{
    Client: redisClient, KeyPrefix: "doc:", Embedding: embedder,
})
ids, _ := indexer.Store(ctx, docs)

// 2. 检索相关文档
retriever, _ := redisRetriever.NewRetriever(ctx, &redisRetriever.RetrieverConfig{
    Client: redisClient, Index: "my_index", Embedding: embedder,
})
docs, _ := retriever.Retrieve(ctx, "user query", retriever.WithTopK(5))

Tool Usage

工具使用

MCP Tools

MCP工具

import mcpp "github.com/cloudwego/eino-ext/components/tool/mcp"

tools, err := mcpp.GetTools(ctx, &mcpp.Config{Cli: mcpClient})

import mcpp "github.com/cloudwego/eino-ext/components/tool/mcp"

tools, err := mcpp.GetTools(ctx, &mcpp.Config{Cli: mcpClient})

Custom InvokableTool

自定义InvokableTool

Implement

Info()

and

InvokableRun()

to create a custom tool.

实现

Info()

和

InvokableRun()

方法即可创建自定义工具。

Instructions to Agent

给Agent的开发说明

Constructor signatures and Config struct names vary across implementations. Always read the provider's reference file in
```
reference/{type}/{impl}.md
```
before generating initialization code.
Use
```
ToolCallingChatModel
```
(not deprecated
```
ChatModel
```
) for tool binding.
For RAG, ensure the same Embedder model is used for both indexing and retrieval.
See reference files for detailed per-component documentation.

不同实现的构造函数签名和Config结构体名称各不相同。生成初始化代码前，请务必阅读
```
reference/{type}/{impl}.md
```
路径下对应提供商的参考文档。
工具绑定时请使用
```
ToolCallingChatModel
```
（而非已废弃的
```
ChatModel
```
）。
实现RAG功能时，请确保索引和检索阶段使用同一个Embedder模型。
各组件的详细文档请参阅参考文件。

Reference Files

参考文件

Read files on-demand for detailed API, config, and examples. Each

{type}/

directory contains an

overview.md

(interfaces + common patterns) and per-implementation files:

```
reference/model/*.md
```
-- ChatModel interfaces, tool binding, streaming, and per-provider config (openai, claude, gemini, ark, ollama, deepseek, qwen, qianfan, openrouter)
```
reference/embedding/*.md
```
-- Embedder interface and per-provider config (openai, ark, ollama, etc.)
```
reference/retriever/*.md
```
-- Retriever interface, RAG example, and per-backend config (redis, milvus2, es8)
```
reference/indexer/*.md
```
-- Indexer interface, indexing pipeline, and per-backend config (redis, milvus2, es8, qdrant)
```
reference/tool/*.md
```
-- Tool interfaces, custom tool creation, MCP integration, search tools, utility tools
```
reference/document/pipeline.md
```
-- Loader, Parser, Transformer interfaces and full pipeline example
```
reference/prompt.md
```
-- ChatTemplate, FString/GoTemplate/Jinja2 formats, message helpers
```
reference/callback/*.md
```
-- Callback handler interface, registration patterns, and per-provider config (cozeloop, apmplus, langfuse, langsmith)

按需阅读对应文件即可获取详细API、配置说明和使用示例。每个

{type}/

目录都包含

overview.md

（接口定义+通用使用模式）和各实现对应的独立文档：

```
reference/model/*.md
```
-- ChatModel接口、工具绑定、流式处理、各提供商配置（openai、claude、gemini、ark、ollama、deepseek、qwen、qianfan、openrouter）
```
reference/embedding/*.md
```
-- Embedder接口和各提供商配置（openai、ark、ollama等）
```
reference/retriever/*.md
```
-- Retriever接口、RAG示例、各后端配置（redis、milvus2、es8）
```
reference/indexer/*.md
```
-- Indexer接口、索引流水线、各后端配置（redis、milvus2、es8、qdrant）
```
reference/tool/*.md
```
-- Tool接口、自定义工具创建、MCP集成、搜索工具、实用工具
```
reference/document/pipeline.md
```
-- Loader、Parser、Transformer接口和完整流水线示例
```
reference/prompt.md
```
-- ChatTemplate、FString/GoTemplate/Jinja2格式、消息辅助工具
```
reference/callback/*.md
```
-- 回调处理器接口、注册模式、各提供商配置（cozeloop、apmplus、langfuse、langsmith）