LLM Models

Lists the LLM models in the current catalog, in the OpenAI list shape. The list is the same for every account.

#Request

curl https://api.sogni.ai/v1/models \
  -H "Authorization: Bearer $SOGNI_API_KEY"

const res = await fetch('https://api.sogni.ai/v1/models', {
  headers: { Authorization: `Bearer ${process.env.SOGNI_API_KEY}` },
});
const { data } = await res.json();

import os
from openai import OpenAI

client = OpenAI(base_url="https://api.sogni.ai/v1", api_key=os.environ["SOGNI_API_KEY"])
models = client.models.list()

#Response

{
  "object": "list",
  "data": [
    {
      "id": "qwen3.6-35b-a3b-gguf-iq4xs",
      "object": "model",
      "created": 1776384000,
      "owned_by": "qwen",
      "capabilities": { "reasoning": true },
      "context_length": 262144,
      "max_model_len": 262144,
      "sogni": {
        "usable_context_length": 260096,
        "safety_margin_tokens": 2048,
        "max_tokens_behavior": "clamp_to_remaining_context"
      }
    },
    {
      "id": "deepseek-v4-flash-vision-exp-dspark-1m",
      "object": "model",
      "created": 1788998400,
      "owned_by": "deepseek",
      "capabilities": { "reasoning": true },
      "context_length": 1048576,
      "max_model_len": 1048576,
      "sogni": {
        "usable_context_length": 1046528,
        "safety_margin_tokens": 2048,
        "max_tokens_behavior": "clamp_to_remaining_context"
      }
    },
    {
      "id": "qwen3.5-35b-a3b-abliterated-gguf-q4km",
      "object": "model",
      "created": 1772582400,
      "owned_by": "qwen",
      "capabilities": { "reasoning": true },
      "context_length": 131072,
      "max_model_len": 131072,
      "sogni": {
        "usable_context_length": 129024,
        "safety_margin_tokens": 2048,
        "max_tokens_behavior": "clamp_to_remaining_context"
      }
    }
  ]
}

#Context and output limits

context_length and max_model_len report the full configured context used by API admission. The sogni.usable_context_length allowance covers prepared messages, tool/response schemas and output together after the sogni.safety_margin_tokens reserve. These fields are additive provider metadata; OpenAI's model object does not standardize context discovery, and some clients require manual limit settings.

For hosted completions, valid positive integer max_tokens or max_completion_tokens ceilings are reduced to the remaining estimated space. Smaller ceilings are preserved; input and caller tools are not shortened. If input alone leaves no room for output, the request returns a context error with numeric budget details. See Context and Output Limits for header settings and estimation limits.

Returns the same model record and context metadata as the list endpoint. An unknown id returns 404 with code: "model_not_found". A retired model id that has a replacement returns the replacement's record.

#Path parameters