Double any deposit — up to $250 in matched credits, through Aug 10 →

Serverless Inference

Run inference without managing a single GPU

An OpenAI-compatible API for chat, image, and video models. Pay per token, scale to zero, and ship in one line of code — no instances to launch or shut down.

$0Cost when idle — scales to zero
<1sTo first token, no cold start
1 keyChat, image, and video in one API
Quickstart

Swap two lines, keep your stack

Because the API is OpenAI-compatible, the official SDKs, curl, and our CLI all work against it out of the box.

from openai import OpenAI

client = OpenAI(
    base_url="https://api.gpu.ai/v1",
    api_key="gpuai_live_...",
)

stream = client.chat.completions.create(
    model="gpuai/qwen2.5-7b-instruct",
    messages=[{"role": "user", "content": "Explain WireGuard in one sentence"}],
    stream=True,
)

for chunk in stream:
    print(chunk.choices[0].delta.content or "", end="")

Get a key with gpu login or mint one in the dashboard. See the full serverless inference guide.

Why Serverless

Inference without the infrastructure

All the capability of dedicated GPUs, none of the fleet management.

OpenAI-compatible

Point any OpenAI SDK, curl, or the gpu CLI at our endpoint. Swap the base URL and key — your existing code works unchanged.

Pay per token

No instances to launch, patch, or shut down. You're billed per token, per image, or per second of video — nothing when you're idle.

No cold starts

The default serverless tier keeps models warm, so requests return in milliseconds. Opt into the economy tier when price matters more than latency.

Live model catalog

The catalog is probe-verified and refreshed hourly from what's actually serving warm right now — GET /v1/models always reflects what you can call this minute.

Autoscaling built in

Capacity flexes with your traffic automatically. Send one request or ten thousand per second — there's no fleet to provision or tune.

Three modalities, one endpoint

Chat completions, image generation, and video generation share the same base URL, the same auth, and the same billing.

One API, three modalities

Chat, image, and video

Same base URL, same key, same billing — pick the modality your product needs.

Chat

Priced per 1M tokens

Streaming chat completions across open-weight LLMs. Drop-in replacement for the chat completions endpoint you already call.

Image

Priced per megapixel

Text-to-image generation from fast, high-quality diffusion models. Submit a prompt, get pixels back.

Video

Priced per output second

Text-to-video generation. Submit a job, poll for progress, and download the finished clip when it's ready.

Pricing

Pay only for what you generate

Representative starting prices — the live catalog surfaces the cheapest partner per model.

ModelModalityFrom
Qwen2.5 7B InstructChat$0.20 / 1M tokens
Llama 3.3 70B InstructChat$0.60 / 1M tokens
FLUX.1 [schnell]Image$0.003 / Mpxl
Wan 2.2 T2VVideo$0.05 / output sec

Pricing updates live across partners. Run gpu models list for current rates, or opt into the :economy tier for cheaper, compute-priced inference. Need a dedicated instance instead? On-Demand gives you the whole GPU.

Ship your first
inference call today