Skip to content

Dedicated inference

A private assistant.
For your business.

Qwen3.8 27B on 1× NVIDIA A100 SXM4, 80 GB in US-MD-1, United States. Draft, research, summarize, and work with business documents through an authenticated API for $1,899 per month per endpoint.

Provisioning begins after verified payment, subject to current qualified capacity. Setup and custom integrations are quoted separately. This is a model API, not a ready-made chat app.

Everyday business work

More than an extraction tool.

  • Draft emails and documents
  • Research and brainstorm
  • Summarize long documents
  • Ask questions over images and PDFs
  • Extract structured JSON, including invoice fields

Up to 32K context tokens · up to 4K output tokens per request. Text, PNG/JPEG/WebP images, and PDFs (scanned or digital, up to 10 pages).

Each endpoint serves up to 8 requests at once; additional requests wait briefly in a bounded queue. This configured limit is not a measured concurrency qualification. Audio, video, and tool calling are not supported. Thinking is disabled. Check important answers against your sources.

See image and PDF API examples and limits

Keep control of your business data

Private by design.

Put data security at the center of your AI setup. Your model runs on a dedicated GPU, separate from other customers’ runtimes. We don’t use your prompts, files, or outputs to train models or target ads.

Review access and data handling

Account services are shared; operator access and retention are described in our security information and agreed terms. Evaluate the model on your own tasks before relying on it: performance and accuracy for this model have not yet been measured by weightroom.

Deploy a model