Skip to content

Pricing

Your workload.
Your compute budget.

Choose a dedicated monthly endpoint or a separate GPU workspace. Review current capacity and terms before checkout.

Available today

GPU workspaces

Rates depend on your GPU, provider, and storage configuration. Current options are shown in the launch flow.

  • Your own model runtime
  • Preconfigured workspace templates
  • Usage and credit balance in your dashboard

Dedicated inference

Qwen3.8 27B · US endpoint

$1,899/ month per endpoint

Qwen3.8 27B on 1× NVIDIA A100 SXM4, 80 GB in US-MD-1, United States. Provisioning begins after verified payment, subject to current qualified capacity. Setup and custom integrations are quoted separately.

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). Draft emails and documents, brainstorm, summarize, ask questions over files, or extract structured JSON.

Each endpoint serves up to 8 requests at once; additional requests wait briefly in a bounded queue. Audio, video, and tool calling are not supported. Thinking is disabled. Review file and request limits.

Subscribe securely with Stripe from your team dashboard. This monthly subscription is separate from Workspace credits and per-minute billing.

Compute and storage are separate charges. Stopping compute can leave storage and backup costs running.

Review the launch summary and retention settings for your workspace. Your dashboard shows usage and available credit.