Keep your familiar workflow
Configure the Dev.dog endpoint, key, and model ID in a tool that supports custom APIs. Explore more models with the tools you already use.
UNIFIED AI MODEL ACCESS & MANAGEMENT
Your next ideabelongs here.
Everyday coding, content creation, or application development can all start here. Access AI models through Dev.dog and focus on what you really want to accomplish.
OpenAI, Anthropic, and Gemini API protocols.
MORE MODELS, MORE CHOICE
Find the right model for your work.
From coding to document processing, compare capabilities and choose for your needs and budget.
Claude
AnthropicWriting, long documents, and complex code analysis.
GPT
OpenAICoding, assistants, and application development.
Gemini
GoogleTasks combining text, images, and documents.
Grok
xAIConversation, reasoning, and problem analysis.
DeepSeek
DeepSeekReasoning, code generation, and Chinese-language tasks.
Qwen
AlibabaChinese, multilingual tasks, and open-model applications.
Kimi
Moonshot AILong documents, research materials, and writing.
Check the catalog and selected group for specific versions, supported features, and prices.
Browse models & group pricingBUILT FOR EVERYDAY USE
From connecting to working with AI.
Keep keys, usage, and spending clear as your projects grow.
Manage each project separately
Create separate keys for everyday tools, testing, and production. Rotate or revoke one without affecting your other projects.
Understand each request’s cost
View requests, token usage, and charges in the console. Compare results and spending to find an approach that works for you.
GET CONNECTED
Make your first request work.
Configure your client’s connection settings, or start with the code example to build your own application.
- 1
Choose a model and group
Sign in, compare capabilities and group prices, and choose a suitable option. Confirm that your account has available credit.
- 2
Create a project key
Create an API key for your chosen group. Enter it in your client, or store it in a server environment variable for your application.
- 3
Configure and verify
Set the endpoint and exact model ID from the guide. Send a short request, confirm the response, and check its usage and charge.
OpenAI-compatible example
import os
from openai import OpenAI
client = OpenAI(
api_key=os.environ["DEV_DOG_KEY"],
base_url="https://dev.dog/v1",
)
response = client.chat.completions.create(
model="YOUR_MODEL_ID",
messages=[{"role": "user",
"content": "Hello!"}],
)
print(response.choices[0].message.content)import OpenAI from "openai";
const client = new OpenAI({
apiKey: process.env.DEV_DOG_KEY,
baseURL: "https://dev.dog/v1",
});
const response = await client.chat.completions.create({
model: "YOUR_MODEL_ID",
messages: [{ role: "user", content: "Hello!" }],
});
console.log(response.choices[0].message.content);curl https://dev.dog/v1/chat/completions \
-H "Authorization: Bearer $DEV_DOG_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "YOUR_MODEL_ID",
"messages": [{
"role": "user",
"content": "Hello!"
}]
}'SDK setup
pip install openaiSet DEV_DOG_KEY to your key and YOUR_MODEL_ID to a model ID supported by your group.
Different tools may expect different endpoint formats. View setup details
BEFORE YOU CONNECT
Frequently asked questions
About tool compatibility, model groups, and usage charges.
Need setup or troubleshooting details?
Read the connection guideCan I connect my existing coding tools and clients?
If your tool supports a custom endpoint, API key, and model name, follow the Dev.dog guide and choose the matching OpenAI, Anthropic, or Gemini protocol. Streaming, tool calls, and image input depend on the model and channel. Verify the features you need before making it part of your daily workflow.
Which endpoint should I use?
For the OpenAI SDK, use https://dev.dog/v1. For Anthropic or Gemini clients, use the corresponding address in the guide. Some tools add /v1 automatically, so check their Base URL requirements to avoid repeating the version prefix.
How do groups for the same model differ?
Groups represent different access channels, with potentially different prices, features, and limits. Compare them in the catalog, then create a key for your chosen group. Test your actual tasks and review their costs before choosing a group for ongoing use.
How are requests billed, and can I see the charges?
The console shows request history and charges. Text models are usually billed by tokens, the units of text a model processes. Input, output, and cached tokens may have different prices. See the selected model and group for exact rates and other billing methods. Start with a short request to check actual usage.
Give your next idea a start.
Choose a model, create a key, and bring AI into your tools and projects.
Get started