# Perceptron: Perceptron Mk1.5

Provider: perceptron  
Category: llm  
Model ID: `perceptron/perceptron-mk1.5`

Perceptron Mk1.5 is Perceptron's embodied reasoning model for physical agents. It accepts text, image, video, and audio input, and answers with text plus optional structured annotations: points, boxes, polygons, tracks,...

## Specs

- Context length: 36864 tokens
- Max output: 8192 tokens
- Modalities: text, image, video, audio
- Released: 2026-09-25

## Pricing

- Input: $0.15 per million tokens
- Output: $1.50 per million tokens

## Providers

- **perceptron** — ctx 36864, input $0.15/M, output $1.50/M

## Verdict

Perceptron Mk1.5 is a multimodal model that handles text, images, video, and audio in a single context window of 36K tokens. At $0.15/$1.50 per Mtok, it sits in the budget tier but offers broader modality support than most peers at this price point. The lack of public benchmarks makes performance hard to gauge against alternatives. Reach for this when you need multimodal flexibility on a tight budget and can validate outputs yourself.

---
Last verified: 2026-10-19T02:30:34.239Z  
Canonical URL: https://switchy.build/directory/models/perceptron-mk1-5