One line of code.
Every major model.
Call Claude, Gemini, GPT, DeepSeek and more through a single endpoint — no per-vendor integration work. Cheaper than going direct, simpler than building it yourself.
from openai import OpenAI
client = OpenAI(
base_url="https://api.eclogue.ai/v1",
api_key="gw-your-key-here",
)
response = client.chat.completions.create(
model="deepseek-chat",
messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)Why Eclogue
Built around the problems that actually slow AI teams down, so you can spend your time on the product instead of the plumbing.
Volume pricing, passed on
We buy in bulk and hand the discount over. Pay for what you actually use, with no hidden fees.
Multiple protocols, natively
Fully compatible with the OpenAI API, plus Anthropic Messages, Gemini, Responses and video generation. Switching over is a two-line config change — no rewrite.
Cache hits are discounted too
When a prompt cache hits, you are billed at the cache rate. Some gateways quietly charge full price for cached tokens; we bill what you used and nothing more.
Real-time cost visibility
Every call is costed to the cent as it happens. Group usage by project or by user, so month-end reconciliation stops being a chore.
Granular access control
Give each team its own key with its own model allow-list and budget. Cheap models in staging, frontier models in production, spend under control.
We do not keep your data
The hosted service does not retain conversation content. Need internal audit trails? A self-hosted deployment is available — one Docker command, and the data never leaves your side.
Major models already available, with more added continuously
Live in five minutes
A two-line change migrates you over, and the savings show up the same day. No credit card required — every new account starts with trial credit.