Models
List the model ids your key can call, and find which route each model family answers on.
Listing models
GET /v1/models returns the catalog. It is answered by Token Share itself, so it is exactly the allowlist requests are checked against: an id in the list is callable, and one that is not will be refused. GET /v1/models/{id} returns a single entry.
curl https://token-share.app/v1/models \
-H "Authorization: Bearer $TOKEN_SHARE_KEY"The response shape follows your headers: send anthropic-version for the Anthropic listing, nothing for the OpenAI one. Copy ids from here rather than from a vendor changelog — they are what the allowlist compares.
Which route a model uses
| Family | Route | Auth header |
|---|---|---|
| Claude | POST /v1/messages | x-api-key |
| GPT | POST /v1/responses | Authorization: Bearer |
| Grok, Gemini | POST /v1/chat/completions | Authorization: Bearer |
| Image models | POST /v1/images/generations | Authorization: Bearer |
Each model page states its exact route, context window and price. Sending a model to the wrong route is rejected before it is forwarded.
Pricing
You pay each model’s official list price times its multiplier, per token. The pricing page and every model page list both; the Billing API returns what each request actually cost.