Quick setup
Open the kRouter dashboard at http://localhost:20128
Go to Providers → click Vertex Partner
Paste your API key, click Save
Test the connection
Use in your IDE
Point any OpenAI-compatible client at kRouter using your local API key.
# Endpoint
http://localhost:20128/v1
# API key (from Dashboard → API Keys)
sk-krouter-XXXX
# Model
vxp/<model-id>How Vertex Partner billing works with kRouter
Vertex Partner is pay-per-token. You bring your own API key and kRouter routes to it, so you keep your own rate limits and billing relationship. kRouter authenticates it via apikey. 4 models are routable as vertex-partner/<model>.
Through kRouter, Vertex Partner serves chat and code completion. Requests reach it through the same local OpenAI-compatible endpoint as every other provider, so switching to or from Vertex Partner is a one-line model change in your client rather than a rewrite.
Vertex Partner models & pricing
Prices are USD per 1M tokens, cheapest first. kRouter bills nothing on top; this is the provider's own rate.
| Model | Route as | Input /1M | Output /1M |
|---|---|---|---|
| DeepSeek V3.2 (Vertex) | vertex-partner/deepseek-ai/deepseek-v3.2-maas | $0.14 | $0.28 |
| Qwen3 Next 80B Thinking (Vertex) | vertex-partner/qwen/qwen3-next-80b-a3b-thinking-maas | $0.50 | $2.00 |
| Qwen3 Next 80B Instruct (Vertex) | vertex-partner/qwen/qwen3-next-80b-a3b-instruct-maas | $0.50 | $2.00 |
| GLM-5 (Vertex) | vertex-partner/zai-org/glm-5-maas | — | — |
Vertex Partner links
Vertex Partner FAQ
Is Vertex Partner free through kRouter?
Vertex Partner is pay-per-token: you bring your own API key and are billed by Vertex Partner at their rates. kRouter adds nothing on top and is free and MIT-licensed.
How do I use Vertex Partner with Claude Code or Cursor?
Connect Vertex Partner in the kRouter dashboard, then point your tool at kRouter's local endpoint (http://localhost:20128/v1) with a kRouter API key. Any OpenAI-compatible client works, and models are addressed as vertex-partner/<model>.
Which Vertex Partner models can kRouter route to?
4 models, including DeepSeek V3.2 (Vertex), Qwen3 Next 80B Thinking (Vertex), Qwen3 Next 80B Instruct (Vertex). Every one is addressed as vertex-partner/<model> from any client.
What happens when Vertex Partner is rate limited?
kRouter detects the limit, parks that account until its real reset time, and fails over to your next connected account or provider — so the request still completes instead of erroring.