vLLM Endpoint

Get an OpenAI-compatible API for any open model

Deploy any supported open-source model behind a standard OpenAI-compatible /v1/chat/completions and /v1/completions API — swap your existing OpenAI SDK base URL and keep your code unchanged.

Run this jobGet help running this

Estimated Varies by model — see the model's own /serve/models page · Long-running endpoint — starts in 2–8 min depending on model size

What inputs you need

  • Any Badgr-supported open model name
  • Max-cost budget
  • Existing OpenAI SDK client (just change the base URL)

What Badgr returns

  • OpenAI-compatible endpoint URL + API key
  • Health-check status
  • Usage receipt on teardown

Recommended GPU routes

H100 80GBA100 80GBL40S 48GB

Estimated cost: Varies by model — see the model's own /serve/models page · Estimated runtime: Long-running endpoint — starts in 2–8 min depending on model size

Example command

badgr serve <model-name> \
  --gpu H100 \
  --max-cost 10

Common setup failures Badgr avoids

Client code assumes a real OpenAI endpoint's rate limits — Badgr's endpoint enforces your own max-cost budget instead, so failures are budget-driven, not silent throttling.

Endpoint fails health check on cold start — startup timeout is sized per model weight size automatically.

Ready to run this?

Launch from the dashboard, CLI, or Compute API. Max-cost protection included.

Run this jobBack to vLLM Endpoint