Perceptron Mk1
Perceptron Mk1 is a large language model made by Perceptron, added to the catalog on 12 May 2026. It has a context window of 32,768 tokens and costs $0.15 per 1M input tokens and $1.50 per 1M output tokens. It accepts text, images and video and produces text. Typical uses: answering questions about images, answering questions about video and describing visual scenes.
Price checked ยท source: OpenRouter
What Perceptron Mk1 is
Perceptron Mk1 is a vision-language model from Perceptron for video and embodied reasoning. It accepts text, image, and video inputs, including images or video paired with natural-language queries, and returns text. Its stated focus is detailed visual understanding, so its role is interpreting visual material in response to a written prompt rather than generating images or video.
The model has a context length of 32,768 tokens and can produce up to 8,192 output tokens. This provides room for prompts that include text alongside visual input, while the output limit allows for responses beyond a short caption. The facts do not specify how video is represented within the context or define supported file formats, so those details depend on the serving interface.
Perceptron Mk1 supports reasoning and structured outputs. These capabilities suit tasks that call for an explanation based on visual input or a response arranged in a defined structure. It may be used for image and video question answering, scene description, and reasoning about visual events when the task is stated in text. It is intended for prompts where visual content is central to the question. The model produces text only; its listed capabilities do not include audio input or generated media.
Key features
- Accepts text, image, and video inputs
- Produces text output
- Has a 32,768-token context length
- Supports reasoning and structured outputs
- Can produce up to 8,192 output tokens
Typical uses
- Answering questions about images
- Answering questions about video
- Describing visual scenes
- Reasoning about visual events
- Producing structured responses to visual prompts
Perceptron Mk1 pricing
US dollars, list price through OpenRouter.
| Item | Standard |
|---|---|
| Input, per 1M tokens | $0.15 |
| Output, per 1M tokens | $1.50 |
No price change since it was added on 12 May 2026.
| What a task costs | Cost |
|---|---|
| Answer a short question500 in / 300 out | $0.0005 |
| Summarise a 50-page PDF25,000 in / 800 out | $0.0050 |
| Review a 2,000-line source file30,000 in / 2,000 out | $0.0075 |
| Read a 300-page contract set150,000 in / 1,500 out | $0.025 |
Perceptron Mk1 specifications
| Model id | perceptron/perceptron-mk1 |
|---|---|
| Version | perceptron-mk1-20260512 |
| Context window | 32,768 tokensAbout 24,576 English words |
| Max output | 8,192 tokens |
| Knowledge cutoff | Not published |
| Added to catalog | 12 May 2026 |
| Input types | text, image, video |
|---|---|
| Output types | text |
| Tool calling | No |
| Structured outputs | Yes |
| Reasoning effort control | No |
| Content moderation | Not applied by provider |
Sample outputs
The same prompts are given to every model, and the unedited output is shown with the date it was produced. Upend does not publish leaderboard scores it did not run.
No output has been recorded for Perceptron Mk1 yet.
Compare Perceptron Mk1
Side by side with other models: prices, context window, what each accepts and supports, and the same four tasks priced on both.