API online — 10 models available

AI inference for
builders.

Drop-in OpenAI replacement. Fast models on GPU, frontier models via Morpheus. No prompt logging. Free tier included.

app.py
from openai import OpenAI

client = OpenAI(
    base_url="https://compute.denariilabs.xyz/v1",
    api_key="sk-dc-..."
)

r = client.chat.completions.create(
    model="qwen2.5-7b",
    messages=[{"role": "user",
               "content": "Hello!"}]
)
print(r.choices[0].message.content)
Fast
GPU-powered models respond in milliseconds. Llama 3B, Qwen 7B, Mistral 7B on dedicated hardware.
Private
Zero prompt storage. No training on your data. We log token counts for billing, nothing else.
Compatible
Works with OpenAI SDK, LangChain, and any tool that supports the OpenAI API format. Change one line.
10 Models GPU + Frontier via Morpheus
Llama 3B Qwen 7B Mistral 7B Llama 70B Qwen 235B Hermes 405B +4 more
Free tier included — no card required

Get started

Sign up to get your API key. $5 free credit included.

or

Already have an account? Sign in

Admin

Create API Key

Give your key a label so you can identify it later.

Model preference
Used in the code snippet below. You can use any model per request.
D
Denarii Compute
Step 1 of 3

What are you building?

Helps us tailor your experience. You can skip this.

Pick your starting model

You can switch anytime. Not sure? Start with Balanced.

Fast

Llama 3B

Fastest responses. Great for prototyping and simple tasks.

Recommended
Balanced

Qwen 7B

Best mix of speed and quality. Most popular choice.

Powerful

Llama 70B

Highest quality reasoning. Best for complex tasks.

You're all set!

Here's your API key. Save it — it's shown only once.

Python
cURL
JavaScript
Spend
$0
Tokens
Requests
Latency
Monthly Budget
00
Chat
Pick a model and start chatting. $5 free credit included.
Usage — 30 days
Avg /dayPeak Trend
Treasury
MOR
Revenue
$0
All-time
Compute cost$0
User balances$0
ETH
Morpheus sessions0
Invite Codes