Reasoning models

Requirements and limitations

Reasoning controls and token accounting vary by model. The capability alone does not establish reasoning-effort parameter support or output quality.

Only active chat models with explicit support are included. Missing capability values remain unknown; top-level values take precedence over nested values. Image input also establishes vision support.

Example request

POST https://api.llm7.io/v1/chat/completions
Authorization: Bearer YOUR_LLM7_API_KEY
Content-Type: application/json

{
  "model": "DeepSeek-V4-Flash-0731",
  "messages": [
    {
      "role": "user",
      "content": "Explain how to verify a multi-step calculation."
    }
  ]
}

Explore related pages