CloudGPU

CloudGPU provides an OpenAI-compatible Chat Completions API at https://cloudgpu.app/v1. One API key gives access to models such as DeepSeek, Kimi, GLM and MiniMax.

Setup

go get github.com/zendev-sh/goai@latest

import "github.com/zendev-sh/goai/provider/cloudgpu"

Set the CLOUDGPU_API_KEY environment variable, or pass WithAPIKey() directly. API docs and key setup: cloudgpu.app/api/docs.

Models

Model IDs are bare names:

• deepseek-v4.1-flash
• deepseek-v4-pro
• kimi-k3
• glm-5.3-flash
• minimax-m3

GET https://cloudgpu.app/v1/models lists the full catalog and needs no API key.

Tested Models

Unit tested (mock HTTP server): deepseek-v4.1-flash, glm-5.3-flash

Usage

model := cloudgpu.Chat("deepseek-v4.1-flash")

result, err := goai.GenerateText(ctx, model, goai.WithPrompt("Hello"))
if err != nil {
    log.Fatal(err)
}
fmt.Println(result.Text)

Options

Option	Type	Description
WithAPIKey(key)	string	Set a static API key
WithTokenSource(ts)	provider.TokenSource	Set a dynamic token source
WithBaseURL(url)	string	Override the default https://cloudgpu.app/v1 endpoint
WithHeaders(h)	map[string]string	Set additional HTTP headers
WithHTTPClient(c)	*http.Client	Set a custom *http.Client

Notes

• Declares text-only chat capability with tool calling.
• Reasoning from earlier assistant turns is sent back as reasoning_content, and streaming requests set stream_options.include_usage so the final chunk carries token usage.
• The package provides Chat only. The catalog also lists image, speech and transcription models, which this provider does not wrap.
• Environment variable CLOUDGPU_BASE_URL can override the default endpoint.