Skip to content

Quick Start

Blades is a multimodal AI Agent framework based on the Go language, supporting custom models, tools, memory, middleware, etc., suitable for multi-turn conversations, chain-of-thought reasoning, and structured output scenarios.

Environment Setup

Ensure you have installed Go 1.20+, then initialize your project and import Blades with the following commands:

cd your-project-name
go mod init your-project-name
go get github.com/go-kratos/blades

First Chat Agent

Below is a complete example of building a simple chat Agent using the OpenAI model:

package main

import (
    "context"
    "log"

    "github.com/go-kratos/blades"
    "github.com/go-kratos/blades/contrib/openai"
    "github.com/openai/openai-go/v2/option"
)

func main() {
    // Create an Agent, specifying the model and model provider
    provider := openai.NewChatProvider(
        openai.WithChatOptions(
            option.WithBaseURL("https://api.deepseek.com"),
            option.WithAPIKey("API-YOUR-KEY"),
        ),
    )
    agent := blades.NewAgent(
        "Blades Agent",
        blades.WithModel("deepseek-chat"),  // or gpt-5, qwen3-max, etc.
        blades.WithProvider(provider),
        blades.WithInstructions("You are a helpful assistant that provides detailed and accurate information."),
    )
    // Input prompt
    prompt := blades.NewPrompt(
        blades.UserMessage("What is the capital of France?"),
    )
    // Execute the Agent
    result, err := agent.Run(context.Background(), prompt)
    if err != nil {
        log.Fatal(err)
    }
    // Output the result
    log.Println(result.Text())
}

Note

If you want to switch to another LLM API, you need to set environment variables, for example:

export OPENAI_BASE_URL=https://api.deepseek.com

You need to set the environment variable OPENAI_API_KEY, for example:

export OPENAI_API_KEY=your-api-key

Common Large Models

Provider	Model	Compatibility
OpenAI	ChatGPT, GPT-4, GPT-3.5, etc…	Alibaba Bailian, DeepSeek
Claude	Claude 3.7 Sonnet	-
Gemini	Gemini 2.5 Pro	-

Core Concepts Overview

Component	Description
Agent	The core of the intelligent agent, responsible for coordinating models, tools, memory, etc.
Tool	External capability plugins (such as calling APIs, querying databases)
Memory	Session memory management, supporting multi-turn context
Runnable	A unified interface for all executable components (Agent, Chain, Model, etc. all implement it)
Middleware	Middleware mechanism for cross-cutting concerns like logging, rate limiting, authentication
ModelProvider	Model adapter (e.g., OpenAI, DeepSeek), providing a unified calling interface

More Examples