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 dialogue, chain-of-thought reasoning, and structured output scenarios.

Environment Setup

Ensure you have Go 1.24+ installed, 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

Prerequisites

Before running, you need to configure environment variables for the model side (different ModelProvider variable names may vary; the key is to ensure the API Key and Base URL are available):

OPENAI_BASE_URL
OPENAI_API_KEY
OPENAI_MODEL

Create an 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() {
    // Configure OpenAI API key and base URL using environment variables:
    model := openai.NewModel(os.Getenv("OPENAI_MODEL"), openai.Config{
        APIKey: os.Getenv("OPENAI_API_KEY"),
    })
    agent, err := blades.NewAgent(
        "Blades Agent",
        blades.WithModel(model),
        blades.WithInstruction("You are a helpful assistant that provides detailed and accurate information."),
    )
    if err != nil {
        log.Fatal(err)
    }
    // Create a new input message
    input := blades.UserMessage("What is the capital of France?") // Create a new input message
    // Run the agent
    runner := blades.NewRunner(agent)
    result, err := runner.Run(context.Background(), input)
    if err != nil {
        log.Fatal(err)
    }
    log.Println(result.Text())
}

Note

If you want to switch to another provider’s 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

Supported Models and Providers

Provider	Models	Compatibility
OpenAI	ChatGPT, gpt-5, gpt-4, etc…	qwen3-mqx、deepseek-chat
Claude	Claude 3.7 Sonnet	-
Gemini	Gemini 2.5 Pro	-

Core Concepts Overview

Component	Description
Agent	Agent interface for integrating and coordinating various functional agents such as models, tools, and memory
Tool	External capability plugins (e.g., calling APIs, querying databases)
Memory	Conversation memory management, supporting multi-turn context
Middleware	Middleware mechanism for cross-cutting concerns such as logging, rate limiting, and authentication
ModelProvider	Model adapter (e.g., OpenAI, DeepSeek), providing a unified calling interface