Integration guide
WorkBuddy on Seedrise.
Add one custom model provider and WorkBuddy can switch between Claude, GPT, Gemini, DeepSeek, Kimi and Qwen on the same key, at wholesale rates, on one invoice.
Why run WorkBuddy through Seedrise
Every model, one provider entry
WorkBuddy's custom model slot points at Seedrise once. From there, any model ID on your rate card works, including Claude through the OpenAI-compatible endpoint.
Wholesale rates
Below the providers' list prices and stepping down with volume. Usage by project on a single USD or AUD invoice.
Australian regions
Pin selected models to Sydney or Melbourne for your keys. Prompts and outputs are never used for training.
Setup in five minutes
- Get your Seedrise key. Request a rate card, receive your console login and create a key for WorkBuddy. Allow the models your team should see.
- Open WorkBuddy's model settings. Choose Custom model with the Chat Completions protocol.
- Fill in the provider. API URL
https://api.seedrise.ai/v1, API key = your Seedrise key, model name = any ID on your rate card, for exampleclaude-sonnet-5-5ordeepseek-v3.2. - Add more models. Repeat the model entry for each ID you want in the picker. Same URL and key each time.
- Save and test. Send a message. The reply should arrive within a few seconds and show up in the Seedrise console under that key.
Check it is working
- The first reply arrives and the Seedrise console shows a request for the model you chose.
- Streaming and tool calls work: Claude models are translated to Anthropic Messages on the gateway, so nothing changes on your side.
- If WorkBuddy offers a connection test, it calls
GET /v1/modelsand should list the models on your key.
For teams
- Create one key per team or department; usage shows per key on the invoice.
- Use the key's model allow-list to decide which models the picker exposes, instead of configuring every seat by hand.
- Set a monthly budget on each key so a runaway workflow stops at a known number.
Troubleshooting and questions
Which URL do I enter, with or without /v1? +
With:
https://api.seedrise.ai/v1. WorkBuddy uses the OpenAI Chat Completions shape, which lives under /v1.Can I use Claude models in WorkBuddy? +
Yes. Enter the Claude ID as the model name; the gateway converts the request to Anthropic Messages and back.
The model name is rejected (400). +
Use the exact ID from the model library, no vendor prefix, and make sure that model is on your key's allow-list (403 means it is not).
Does it work for teams outside Australia? +
Yes. Seedrise serves global teams; billing is in USD or AUD and Australian regions are optional.
Where do I see spend? +
In the Seedrise console per key and model, and itemised on the monthly invoice.
Ready to switch WorkBuddy to wholesale rates?
Per-model rates in writing within one business day, then a console login and a trial key.
Also: full docs · model library · security and data