> ## Documentation Index
> Fetch the complete documentation index at: https://docs.perceptron.inc/llms.txt
> Use this file to discover all available pages before exploring further.

# Perceptron Mk1 API reference

> Endpoint pages for Perceptron Mk1 and how its requests differ from Mk1.5.

This is the API reference for **Perceptron Mk1**. The pages here describe the request shape Mk1 accepts.

All endpoints use the base URL `https://api.perceptron.inc` and take your API key in the `Authorization: Bearer $PERCEPTRON_API_KEY` header; see [Authentication](https://docs.perceptron.inc/perceptron-mk1.5/guides/python/auth.md).

## What differs from Mk1.5

* Mk1 uses the top-level `vision_config` object (`annotation_format`, `enable_thinking`) for grounding and thinking; see [Chat completions](https://docs.perceptron.inc/perceptron-mk1/api-reference/endpoint/chat-completions.md).
* Mk1 does not accept `tools`, `tool_choice` or `parallel_tool_calls`. The full gateway request schema is in the [Perceptron Mk1.5 reference](https://docs.perceptron.inc/api-reference/chat-completions/create-chat-completion.md).
* Multilook on Mk1 does not support `stream`, `response_format`, `stop` or tools; see [Multilook chat completions](https://docs.perceptron.inc/perceptron-mk1/api-reference/endpoint/multilook-chat-completions.md).

> [card] [Perceptron Mk1.5 reference](https://docs.perceptron.inc/api-reference/index.md)
>
> The live contract for the latest model, generated from the OpenAPI spec.

> [card] [Services](https://docs.perceptron.inc/perceptron-mk1/api-reference/services/detect-objects.md)
>
> Detect and Egocentric endpoints.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.