Models

GET https://app.syntic.ai/v1/models
GET https://app.syntic.ai/v1/models/{model_id}

List the models your key can use, with their limits, capabilities and prices, or fetch one model by id. Use it to fill a model picker, to read the context window instead of hard-coding it, and to confirm a model offers a feature (images, reasoning, sampling settings) before you send a request.

Today the catalog is published only on Models overview and Pricing.

Authentication

Authorization: Bearer sk_live_... (any inference key). The list is filtered to what that key’s organisation and workspace may use. Admin keys may call it too.

List models

GET /v1/models

The response is the OpenAI-compatible list (object: "list", data[] of object: "model"). Syntic adds fields on top of it, described in Syntic extensions; clients that only know the standard list ignore them.

{
  "object": "list",
  "data": [
    {
      "id": "Amara S1.2",
      "object": "model",
      "display_name": "Amara S1.2",
      "family": "amara",
      "description": "General work: drafting, analysis, coding help, conversation.",
      "default": true,
      "available": true,
      "context_window": 262144,
      "max_output_tokens": 32768,
      "capabilities": {
        "tools": true,
        "parallel_tool_calls": true,
        "strict_tools": true,
        "json_object": true,
        "json_schema": true,
        "vision": true,
        "documents": true,
        "citations": false,
        "reasoning": false,
        "temperature": true,
        "top_p": true,
        "top_k": false,
        "penalties": false,
        "seed": false,
        "server_tools": ["syntic_web_search", "syntic_web_fetch"]
      },
      "pricing": {
        "currency": "USD",
        "unit": "1M tokens",
        "input": 2.50,
        "cached_input": 0.40,
        "output": 12.00,
        "batch_discount": 0.5
      },
      "zero_data_retention": true,
      "created": 1767225600,
      "deprecation": null
    }
  ]
}

Get one model

GET /v1/models/{model_id} returns a single model object. model_id is the model’s id (its exact display name), URL-encoded (Amara%20S1.2).

The model object

The response never names the company or model behind a Syntic model. id and display_name are Syntic names only.

Syntic extensions

The list and model objects follow the OpenAI-compatible shape (id, object, created). Syntic adds:

curl "https://app.syntic.ai/v1/models?surface=cli" -H "Authorization: Bearer $SYNTIC_API_KEY"

{
  "object": "list",
  "data": [
    { "id": "Amara S1.2", "object": "model", "created": 1767225600, "display_name": "Amara S1.2", "context_window": 262144, "max_output_tokens": 32768, "default": true }
  ]
}

Examples

curl https://app.syntic.ai/v1/models -H "Authorization: Bearer $SYNTIC_API_KEY"
 
curl "https://app.syntic.ai/v1/models/Amara%20Quick" -H "Authorization: Bearer $SYNTIC_API_KEY"

const resp = await fetch('https://app.syntic.ai/v1/models?capability=vision', {
  headers: { Authorization: `Bearer ${process.env.SYNTIC_API_KEY}` },
})
const { data } = await resp.json()
for (const m of data) {
  console.log(`${m.display_name.padEnd(18)} ${m.context_window} ctx  $${m.pricing.input}/$${m.pricing.output}`)
}

import os, requests
 
resp = requests.get(
    "https://app.syntic.ai/v1/models/Amara%20S1.2",
    headers={"Authorization": f"Bearer {os.environ['SYNTIC_API_KEY']}"},
    timeout=10,
)
model = resp.json()
budget = model["context_window"] - model["max_output_tokens"]
print(f"{model['display_name']}: keep prompts under {budget:,} tokens")

Errors

Unified envelope.

Caching and limits

• Responses carry Cache-Control: private, max-age=300. Cache the list for a few minutes; it changes rarely.
• Rate limit: 60 requests per minute per key, separate from inference.

Build notes for the platform team

• Serve the list from the model catalog, including the capability and limit columns.
• Syntic Code (the CLI) is waiting on context_window and max_output_tokens to remove its own copy.
• Never include upstream provider ids in any field, including description.

Related

• Models overview
• Pricing
• Sampling