Run any open-source LLM through an OpenAI-compatible API. Pay per second, not per hour. Start with 50 free calls/day.
openai.ChatCompletionpip install aethonYour GPU sits idle 20+ hours a day. Aethon puts it to work running AI inference and pays you per second of compute time.
We charge for GPU time. That's it. No markup on tokens, no hidden fees. Compare us to anyone.
| Provider | Model / GPU | Per 1K tokens | Hourly rate | Billing |
|---|---|---|---|---|
| OpenAI | GPT-4o mini | $0.015 | — | Per-token only |
| Together.ai | Llama 3.1 8B | $0.0002 | — | Per-token |
| Vast.ai | RTX 4090 | ~$0.0003 | $0.35/hr | Hourly rental |
| Aethonyou are here | RTX 4090 | ~$0.0002 | $0.24/hr | Per-second billing |
Sign up, get your key instantly. 50 free calls/day, no card required.
Set base_url to api.aethon.ai/v1. Your existing OpenAI code just works.
Access 30+ open-source models at a fraction of the cost. Scale as you grow.
Download and run. Auto-detects your GPU, registers with the network.
Enable Gamer Mode, set temp limits. The agent handles everything.
Get paid per second of compute. Cash out via Stripe Connect.
Pay only for GPU time used. No hourly minimums, no wasted spend. 30-second job? Pay for 30 seconds.
Change one line of code. Our API is a drop-in replacement for OpenAI's /v1/chat/completions endpoint.
Auto-pauses when you launch a game, resumes when you're done. Your GPU earns while you sleep.
Temperature monitoring, VRAM guards, automatic throttling. We protect your GPU like it's ours.
Bronze → Silver → Gold. Higher ranks unlock lower platform fees. Long-term suppliers earn more.
50 slots with guaranteed minimum payouts. $15-40/mo based on GPU. Limited time, first come first served.
Be one of our first 50 suppliers. We guarantee a minimum monthly payout regardless of platform demand. Zero risk for you.
6-month commitment. Requires 90%+ uptime. Earn more if demand exceeds the guarantee.
Founders badge is permanent — even after the program ends.
Install the SDK, paste this, run it. That's the whole integration.
$ pip install aethon
# That's it. Works with the OpenAI SDK too.
from openai import OpenAI
client = OpenAI(
base_url="https://api.aethon.ai/v1",
api_key="your-aethon-key"
)
response = client.chat.completions.create(
model="llama3.1:8b",
messages=[
{"role": "user", "content": "Explain GPU compute in one sentence"}
],
stream=True
)
for chunk in response:
if chunk.choices[0].delta.content:
print(chunk.choices[0].delta.content, end="")