Models
Capability Matrix
How to read current Xenodia model capabilities, pricing, modalities, availability, and task support from the live catalog.
Use the live GET /v1/models response as the source of truth. Model IDs, availability, pricing, parameters, and limits can change without a documentation release.
Abridged text-model item
{
"id": "gpt-6-astra",
"object": "model",
"name": "gpt-6-astra",
"modalities": ["text"],
"price_input_per_m": 1041312,
"price_output_per_m": 5206559,
"capabilities": {
"pricing": {
"mode": "tokens",
"unit": "million_tokens"
}
},
"enabled": true,
"available": true
}This is an abridged response example, not a price quote. Monetary fields use micro-USDC units. Read the current item before displaying prices or sending a paid request.
Capability groups
| Group | Examples |
|---|---|
| Text | Chat completions, streaming, tools, reasoning controls. |
| Image | Prompt-only image generation, reference image generation, output formats. |
| Video | Text-to-video, image-to-video, task polling, generation duration. |
| Audio | Speech, transcription, translation, voice controls. |
| Async | Task mode support and polling expectations. |
Client rules
- Render model names, availability, pricing, and supported parameters from
/v1/models. - Use only entries whose
enabledandavailablefields are true. - Treat image and video generation as asynchronous task workflows, even when a legacy
asyncinput is accepted. - Do not infer capabilities from a model name or preserve copied prices in application code.