SDKv2.0.0

Go SDK

Drop-in OpenAI-compatible client for Go. One BaseURL override and you're routing through Meridian with full observability, rate limiting, and cost controls.

go-openai on pkg.go.devGitHub

Quickstart

Install go-openai and point the BaseURL at Meridian. Everything else stays the same.

Terminal
go get github.com/sashabaranov/go-openai

Usage

The only change from a standard go-openai setup is the BaseURL field in the client config.

main.go
package main

import (
    "context"
    "fmt"
    "os"

    openai "github.com/sashabaranov/go-openai"
)

func main() {
    client := openai.NewClientWithConfig(openai.ClientConfig{
        BaseURL: "https://api.getnimbus.net/v1",
        APIType: openai.APITypeOpenAI,
    })

    resp, err := client.CreateChatCompletion(
        context.Background(),
        openai.ChatCompletionRequest{
            Model: "gpt-4o",
            Messages: []openai.ChatCompletionMessage{
                {Role: "user", Content: "Hello from Go!"},
            },
        },
    )
    if err != nil {
        fmt.Fprintf(os.Stderr, "error: %v\n", err)
        os.Exit(1)
    }

    fmt.Println(resp.Choices[0].Message.Content)
}

Streaming

Streaming works identically — set Stream: true and iterate the channel.

main.go
stream, err := client.CreateChatCompletionStream(
    ctx,
    openai.ChatCompletionRequest{
        Model:  "gpt-4o",
        Stream: true,
        Messages: []openai.ChatCompletionMessage{
            {Role: "user", Content: "Tell me a short story."},
        },
    },
)
if err != nil {
    panic(err)
}
defer stream.Close()

for {
    resp, err := stream.Recv()
    if err != nil {
        break
    }
    fmt.Print(resp.Choices[0].Delta.Content)
}

Authentication

Meridian uses the standard Authorization: Bearer header. Set your API key via the config or environment variable.

Config field

client := openai.NewClientWithConfig(openai.ClientConfig{
    BaseURL: "https://api.getnimbus.net/v1",
    APIType: openai.APITypeOpenAI,
    APIKey:  "nimbus_sk_...",
})

Environment variable

export OPENAI_API_KEY="nimbus_sk_..."
# go-openai reads this automatically

Supported Endpoints

Any go-openai method that hits /v1/chat/completions or /v1/embeddings works through Meridian.

CreateChatCompletion

Single-turn chat

CreateChatCompletionStream

Streaming chat

CreateEmbeddings

Text embeddings

ListModels

Available models