Custom Models

Use OpenRouter, Ollama, vLLM, or any OpenAI-compatible endpoint.

Overview

ntrp supports any OpenAI-compatible API endpoint. Register custom models via ~/.ntrp/models.json or the /add-model skill.

models.json

Create ~/.ntrp/models.json with an array of model definitions:

[
  {
    "id": "deepseek-r1",
    "provider": "custom",
    "api_base": "https://openrouter.ai/api/v1",
    "api_key_env": "OPENROUTER_API_KEY",
    "max_output_tokens": 8192,
    "max_context_tokens": 64000
  },
  {
    "id": "llama-local",
    "provider": "custom",
    "api_base": "http://localhost:11434/v1",
    "max_output_tokens": 4096,
    "max_context_tokens": 32000
  }
]

Fields

Field	Required	Description
id	Yes	Model identifier (used in config)
provider	Yes	Must be "custom"
api_base	Yes	Base URL of the OpenAI-compatible API
api_key_env	No	Env var name containing the API key
max_output_tokens	Yes	Maximum output tokens
max_context_tokens	Yes	Maximum context window

Using custom models

After adding to models.json, set the model in your environment or settings:

export NTRP_CHAT_MODEL=deepseek-r1

Or change it in the TUI via /settings.

Providers

OpenRouter

{
  "id": "openrouter/deepseek-r1",
  "provider": "custom",
  "api_base": "https://openrouter.ai/api/v1",
  "api_key_env": "OPENROUTER_API_KEY",
  "max_output_tokens": 8192,
  "max_context_tokens": 64000
}

Ollama

{
  "id": "llama3",
  "provider": "custom",
  "api_base": "http://localhost:11434/v1",
  "max_output_tokens": 4096,
  "max_context_tokens": 128000
}

No api_key_env needed for local Ollama.

vLLM

{
  "id": "my-model",
  "provider": "custom",
  "api_base": "http://localhost:8080/v1",
  "max_output_tokens": 4096,
  "max_context_tokens": 32000
}

Interactive setup

Use the /add-model skill in chat for guided model registration:

/add-model

The agent will walk you through configuring the endpoint.