API online — 10 models available

AI inference for builders.
No OpenAI bill.

Drop-in OpenAI replacement. Fast open-source models on GPU, frontier models via Morpheus. No prompt logging. $5 free to start.

Prototype with 10 models. Ship with the one that works. $5 free credit.

  • OpenAI-compatible — drop-in replacement
  • Streaming, tool use, long context
  • Genesis pricing locked for first 50 users
from openai import OpenAI

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

res = client.chat.completions.create(
    model="qwen2.5-7b",
    messages=[{"role": "user", "content": "Hello"}]
)
print(res.choices[0].message.content)

Fast

GPU-powered open 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 and nothing else.

OpenAI-compatible

Works with OpenAI SDK, LangChain, and any client that speaks the OpenAI format. Change one line.

Ten models.
Fast open-source → frontier reasoning.

Fast tier runs on our GPUs. Frontier and special tiers route through Morpheus — we earn MOR on every request.

Priced for production.

5–10× cheaper than frontier APIs for workloads that don't need frontier intelligence.

Community

Free

Free tier to try it out. No card required.

  • 2M tokens / mo
  • 10 req/min
  • All open models
Most popular

Builder

$50/mo

For individual builders shipping something real.

  • 25M tokens / mo
  • 60 req/min
  • Frontier models via Morpheus
  • Usage dashboard + keys

Founder

$200/mo

For teams with production workloads.

  • 150M tokens / mo
  • 200 req/min
  • Priority GPU routing
  • Email support
See all tiers & pay-as-you-go
Denarii Accelerator

Building in a chain ecosystem?

Hedera, Starknet, Canton, Morpheus — chain-sponsored builders get up to 500M tokens/month covered, plus accelerator mentorship and a direct path to RedBeard Ventures.

Get an API key in 30 seconds.

2.0M free tokens. No card required.