Base URLs & Routing

The Pawan.Krd API supports three URL patterns. They all reach the same endpoints — pick whichever fits your client best.

1. Standard base URL

The classic OpenAI-style base URL. The model is selected via the request bodymodel field.

📄Standard

https://api.pawan.krd/v1

2. Model-scoped base URL (single segment)

Insert the model id directly in the URL. Useful for tools that only let you set a base URL and not a model name (e.g. SillyTavern, Open WebUI, Janitor proxies).

📄Model in path

https://api.pawan.krd/pkrd/cosmosrp-3.5/v1

3. Provider/model base URL (two segments)

For models whose id contains a provider/name namespace (e.g. openai/gpt-oss-20b), you can split the namespace across path segments.

📄Provider + model in path

https://api.pawan.krd/openai/gpt-oss-20b/v1

ℹ

Path overrides body

When a model is present in the URL path, it overrides the model field in the request body. You can omit model from the body entirely.

Endpoints under any base

Every endpoint is reachable from every base URL pattern. For example:

POST /v1/chat/completions
POST /v1/completions
POST /v1/embeddings
POST /v1/rerank
POST /v1/moderations
POST /v1/proofreadings
POST /v1/images/generations, /v1/images/edits, /v1/images/upscale
POST /v1/videos/generations, /v1/videos/edits
POST /v1/audio/speech, /v1/audio/transcriptions, /v1/audio/translations
GET /v1/models, GET /v1/models/{modelId}

Example

💻curl

curl https://api.pawan.krd/openai/gpt-oss-20b/v1/chat/completions \
  -H "Authorization: Bearer pk-your_key" \
  -H "Content-Type: application/json" \
  -d '{
    "messages": [{ "role": "user", "content": "Hello!" }]
  }'

Models & Types

Response Format