Your tools.
Your workflow.
Use the OpenAI SDK or an HTTP client for chat completions, streaming and embeddings.
https://vergeapi.com/v1
The OpenAI-compatible AI gateway
One AI API.
Build with your choice.
Build with GPT, Claude, Gemini, DeepSeek and more through one OpenAI-compatible API. Compare models, set spending limits and track usage in one place.
Model usage is billed at listed rates. Online credit purchases are currently unavailable.
One connection. More choice.Verge API
Your existing SDKPython
client = OpenAI(
base_url="https://vergeapi.com/v1",
api_key=os.environ["VERGE_API_KEY"]
)Choose a model for each request.Chat · Vision · Embeddings
Built around the models you already use
OpenAIAnthropicGoogleMetaDeepSeekMistral
The model catalogue
Compare AI models.
Choose what fits.
Compare models for chat, reasoning, coding, vision and embeddings. Find current prices and capabilities in the live catalogue.
OpenAI
GPT-4.1 Mini
Text and image input with tool calling and a long context window.
openai/gpt-4.1-mini
Anthropic
Claude Sonnet 4.5
A model for reasoning, coding and tool-based workflows.
anthropic/claude-sonnet-4.5
Google
Gemini 2.5 Flash
Multimodal input with support for reasoning and tools.
google/gemini-2.5-flash
DeepSeek
DeepSeek V3.2
Text generation and tool calling through the same API.
deepseek/deepseek-v3.2
Meta
Llama 4 Maverick
Text and image input for conversational applications.
meta-llama/llama-4-maverick
Mistral
Mistral Small 3.2
A compact model with text, image and tool support.
mistralai/mistral-small-3.2-24b-instruct
Availability and rates are shown in the live catalogue.Explore all available models
Built for the work around AI
One integration.
Control where it matters.
Your tools.
Separate keys.
Clear limits.
Give each application its own API key, expiry date, quota and model permissions.
Verge API key
Know what
your app uses.
Review requests, token usage and charges together in your Verge console.
Open your consoleA familiar place to start
From account
to first request.
- Create your Verge accountSign up and open your console.
- Create an application keySet the access and quota your application needs. Your account must have credit for paid requests.
- Choose a model and connectKeep your key on your server and use a model ID from the catalogue.
Request examplePOST
Python · OpenAI SDK
import os
from openai import OpenAI
client = OpenAI(
base_url="https://vergeapi.com/v1",
api_key=os.environ["VERGE_API_KEY"]
)
response = client.chat.completions.create(
model="openai/gpt-4.1-mini",
messages=[{"role":"user","content":"Hello, Verge!"}]
)
print(response.choices[0].message.content)cURL · Chat completions
curl https://vergeapi.com/v1/chat/completions \
-H "Authorization: Bearer $VERGE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "openai/gpt-4.1-mini",
"messages": [{"role":"user","content":"Hello, Verge!"}],
"max_tokens": 128
}'Set
VERGE_API_KEY as a server environment variable.Good to know
The details,
made simple.
Practical answers to get you from choosing a model to your first request.
Explore the API guideWhat does Verge charge?
Planned credit-purchase fee: 2.5% base fee plus transaction fees. Model usage is billed separately at the rates shown in Models and pricing.
Can I keep using the OpenAI SDK?
Yes. Set the base URL to https://vergeapi.com/v1, create a Verge API key and choose a model ID from the catalogue. Keep the key on your server.
How do I choose a model?
Compare models, capabilities and current rates in Models. Choose the exact model ID that fits your request. Available inputs and tools vary by model.
Can different apps have different limits?
Yes. Create separate API keys with their own quotas, expiry and model permissions. Key limits control spending but do not add credit to your account.
Where do I see usage and charges?
See request charges in Usage Logs, usage trends on Dashboard, and your available account credit in Wallet.
Your next build starts here.
One API for your models. Compare live rates and manage your usage.