Hello from Blade.

Blade is AI infrastructure built for developers by Shadow. Models gives you access to open-weight models through an OpenAI-compatible API, and AI Fabric lets you deploy your own models from Python on managed GPUs in Europe.

What would you like to build?

Models

Use the OpenAI SDK you know.

Choose from 70 open-weight models to generate text and images, transcribe audio or read documents.

  • Qwen
  • Gemma
  • Mistral
  • gpt-oss
  • Whisper
  • DeepSeek OCR
  • Nemotron
  • FLUX
See the Python example
python
import osfrom openai import OpenAIclient = OpenAI(  base_url=    "https://api.models.blade.sh/v1",  api_key=os.environ["BLADE_API_KEY"],)

AI Fabric

Run your own code on GPUs.

Deploy from Python and choose the GPUs your workload needs. Your deployment can scale to zero between jobs.

  • RTX A4500, 20 GB
  • RTX 2000 Ada, 16 GB
  • Up to 8 GPUs per container
  • Pay per running minute
See the deployment commands
bash
$ pip install bladekit$ blade deploy app.py

The team behind Shadow brings ten years of cloud experience to Blade.

Talk to the Blade team.

Tell us about your project. We can help you choose a model or work out how many GPUs you need.

Let’s talk

Contact the team

Loading the form…