ToRouter DocsBrowse the model catalog
Discover every model ToRouter can route to, filter by modality, and open the in-console detail view.
The catalog page is the canonical list of models the gateway can serve. It lives at portal.torouter.ai/models.
Find a model
Open Models in the portal sidebar.
Type a name, vendor or model id into the search box — gpt-5, claude, gemini-2.5, deepseek all work.
Filter by modality. The selector covers Text, Multimodal, Image, Video, Audio, Embedding, Rerank, Other.
Click a card to open the detail page at /models/<slug> for capabilities, notes, and the live sheet in the console.
What a card shows
Each card surfaces what matters for picking a model:
- Display name and vendor logo
- Modality badge
- Context window (when known)
- "From" price — a starting rate on the card for quick comparison
- A Try button that opens the model in the in-portal Playground
"From" figures are approximate. Your statement follows logged usage. Open the model detail page for the live sheet, and read Channel routing to see how requests are assigned.
Availability vs catalog
The catalog lists every model the platform knows about. Whether your key can call a model depends on your account's subscriptions and access in the console. To see only models you can invoke today, open Available channels.
Model detail page
Opening a card takes you to /models/<slug>, which shows:
- Line items, supported endpoints, vendor notes, and other fields as shown for your account in the console
- Supported endpoints (
/v1/chat/completions,/v1/messages,/v1beta/...) - Vendor notes, deprecation status, max output tokens
- Direct link to the Playground pre-filled with that model id