How to Use Perceptron Mk1.5 in ChatGPT
Perceptron Mk1.5 turns images and recordings into descriptions, extracted text, and spatial or timed annotations. Compared with Perceptron Mk1, it adds native audio support and video object tracking.
Prompts to try with Perceptron Mk1.5 in ChatGPT
Copy an example into ChatGPT, or ask in your own words.
Use Perceptron Mk1.5 to read this photographed document. Preserve headings and reading order, and mark text you cannot read.
Use Perceptron Mk1.5 to count the visible mugs in this image. Exclude bowls and explain which objects you counted.
Use Perceptron Mk1.5 to find each time the door opens in this video. Give a separate start and end time for every occurrence.
Use Perceptron Mk1.5 to summarize the decisions and open questions in this five-minute recording. Flag unclear speech.
Use Perceptron Mk1.5 to describe this photo in two sentences. Cover the main objects and their arrangement, separating observations from guesses.
How to set up Perceptron Mk1.5 in ChatGPT
Pick the way that suits you. Each takes about two minutes.
Use the ChatGPT connector
- Add a custom MCP server
In ChatGPT open Plugins, then Add › Add custom MCP server. Name it Spiral, paste this URL and keep Authentication on OAuth. Tick I understand and want to continue, then click Create.
https://api.spiral.ai/mcp
- Click Connect
ChatGPT asks to connect Spiral. Connect takes you to Spiral in your browser.
- Sign in to Spiral and approve
Sign in, pick the organization ChatGPT should use and click Authorize. Spiral gives the connection its own API key; ChatGPT never sees it, and you can remove the connection under Settings › Connected apps.
Use the Spiral skill
- Send ChatGPT this message
Your agent reads Spiral's skill and sets itself up.
set up https://spiral.ai/SKILL.md
- Sign in or paste a key
Approve Spiral in the browser, or give the agent a key from the console.
Use the Spiral CLI
- Install and sign in
Installs the Spiral CLI, signs you in and adds its skill to your coding agents.
npx -y spiral-cli@latest init
Call the Spiral API
- Create an API key
In the console, under API keys.
- Run Perceptron Mk1.5
One request; the result comes back by polling or a webhook.
curl https://api.spiral.ai/v1/runs \ -H "Authorization: Bearer $SPIRAL_API_KEY" \ -H "Content-Type: application/json" \ -H "Idempotency-Key: $(uuidgen)" \ -d '{ "provider": "openrouter", "model": "perceptron/perceptron-mk1.5", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: ChatGPT finds Perceptron Mk1.5 on Spiral, shows the price and runs it.
Perceptron Mk1.5 providers and prices
Spiral runs Perceptron Mk1.5 through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.007995 per run at the default inputs. $0.195 per 1M input tokens, $1.95 per 1M output tokens.
Settings you can ask for
Ask ChatGPT for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Top k | a whole number from 0 | ||
| Top p | 0 to 1 | ||
| Max tokens | 1 to 8192 | The most output tokens to generate. | |
| Temperature | 0 to 2 | ||
| Presence penalty | -2 to 2 | ||
| Reasoning effort | none, minimal, low, medium, or high | high | How much the model reasons. |
| Frequency penalty | -2 to 2 |
Every API field of Perceptron: Perceptron Mk1.5 is on its tool page: open its card above.
What can you do with Perceptron Mk1.5 in ChatGPT?
Ask specific questions about photos and recordings and get descriptions or timed evidence to review. Perceptron Mk1.5 can turn those observations into a table, transcript, or concise summary.
Turn document images into editable text
Transcribe scanned pages in reading order, preserving headings and tables. Mark unreadable text instead of filling it in.
Count visible inventory
Count a named object category in a shelf photo. Define exclusions and request a location for each counted object so the result can be reviewed.
Find repeated actions in video
Locate each door opening or other clearly described event. Return separate start and end times for review.
Review a ball's movement
Follow a clearly identified ball through a sports clip. Request observed positions with timestamps and leave gaps when it disappears.
Describe image collections
Draft captions covering visible objects, their arrangement, and the setting. Keep uncertain interpretations separate from observations.
Summarize recorded discussions
Extract stated decisions and unresolved questions from short recordings. Flag unclear speech rather than inventing missing words.
More on Perceptron Mk1.5 and ChatGPT
Overview
Perceptron Mk1.5 in other agents
More ChatGPT guides
Questions about Perceptron Mk1.5 in ChatGPT
Can ChatGPT use Perceptron Mk1.5?
Yes. Add Spiral to ChatGPT as an MCP server, and ChatGPT can find and run Perceptron Mk1.5 like any built-in tool.
How much does Perceptron Mk1.5 cost in ChatGPT?
Up to $0.007995 per run at the default inputs. $0.195 per 1M input tokens, $1.95 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your ChatGPT plan is separate.
Do I need an OpenRouter account to use Perceptron Mk1.5 in ChatGPT?
No. Spiral runs Perceptron Mk1.5 on its own OpenRouter account. You sign in to Spiral once and never paste a key.
How long does it take to set up Perceptron Mk1.5 in ChatGPT?
About two minutes: add a custom MCP server, then click Connect, then sign in to Spiral and approve.
Does ChatGPT wait while Perceptron Mk1.5 runs?
Perceptron Mk1.5 runs in the background. ChatGPT starts the run, checks on it and gives you the text, json when it's ready.
What is Perceptron Mk1.5 useful for?
Use it to inspect visual evidence: read text in images, count visible objects, locate events in video, and follow an object over time. It also summarizes recordings and transcribes speech.
Does Perceptron Mk1.5 generate audio?
It analyzes recordings and returns text. It does not generate audio.
How much content can Perceptron Mk1.5 handle?
The context window is 36,864 tokens, with up to 8,192 completion tokens. Audio also has a separate cap of 16,384 audio tokens per recording, approximately 21.8 minutes; split longer recordings.
Can Perceptron Mk1.5 track an object through every video frame?
It returns timestamped observations that may be sparse. Treat gaps as missing evidence, especially when the object is hidden or off-screen.
Does valid JSON mean the extracted information is correct?
No. A schema controls the answer's structure, but the model can still misread the image or interpret it incorrectly. Check extracted values against the source.
Start using Perceptron Mk1.5 in ChatGPT
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into ChatGPT:
set up https://spiral.ai/SKILL.md


