How to Use GPT-5 Nano in Amp
GPT-5 Nano handles summaries, classification, and routine text extraction. It scored below GPT-5 Mini on OpenAI's long-context retrieval evaluation, so keep tasks focused and check extracted details.
Prompts to try with GPT-5 Nano in Amp
Copy an example into Amp, or ask in your own words.
Use GPT-5 Nano with minimal reasoning to classify the issue descriptions in this project's issue export as bug, feature request, documentation, or other. Return each issue ID, one label, and a supporting phrase.
Use GPT-5 Nano to summarize the latest test failure log in this project in at most 150 words. List failing test names and reported errors; keep proposed causes separate from observed failures.
Use GPT-5 Nano to turn this project's supplied change descriptions into release notes under Fixes, Features, and Documentation. Keep the result under 250 words and preserve issue references.
Use GPT-5 Nano with minimal reasoning to extract TODO comments from the source passages I select. Return JSON records containing the file path, exact comment, and any explicitly named owner; use null when no owner is stated.
Use GPT-5 Nano to rewrite the selected README section for someone using the project for the first time. Keep every command, path, and version unchanged, and limit the explanation to 200 words.
How to set up GPT-5 Nano in Amp
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- Add the server
No header and no client ID or secret: Amp registers itself with Spiral.
amp mcp add spiral https://api.spiral.ai/mcp
- Sign in
Start amp. Amp opens Spiral's sign-in page in your browser: sign in, pick the organization and click Authorize. No API key.
- Check it
amp mcp doctor
Use the Spiral skill
- Send Amp 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 GPT-5 Nano
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": "openai", "model": "openai/gpt-5-nano", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Amp finds GPT-5 Nano on Spiral, shows the price and runs it.
GPT-5 Nano providers and prices
Spiral runs GPT-5 Nano through OpenAI. Open a card for today's price, with every fee included.
Up to $0.002134 per run at the default inputs. $0.065 per 1M input tokens, $0.0065 per 1M cached input tokens, $0.52 per 1M output tokens.
Settings you can ask for
Ask Amp for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Seed | a whole number | ||
| Max tokens | 16 to 128000 | The most output tokens to generate. | |
| Parallel tool calls | on or off |
Every API field of GPT-5 Nano is on its tool page: open its card above.
What can you do with GPT-5 Nano in Amp?
GPT-5 Nano can turn project issue text, logs, and change descriptions into concise summaries and labels. Use it for focused text tasks that help you decide what to inspect next.
Support ticket classification
Assign incoming ticket text to billing, account access, bug report, or other. Define each label and request a supporting phrase for review.
Meeting recaps
Turn a supplied meeting transcript into a short recap with decisions and action items. Keep owners and deadlines tied to what participants actually said.
Order detail extraction
Extract order numbers, product names, and requested actions from customer messages into consistent records. Represent missing details explicitly.
Customer feedback labels
Classify reviews by sentiment and product topic. Preserve review IDs so people can trace each label to the original text.
Release note summaries
Condense supplied change descriptions into short release notes organized by fixes, features, and documentation.
Plain-language rewrites
Rewrite short notices for easier reading while preserving dates, names, and stated conditions. Specify the audience and target length.
Developer issue triage
Classify issue descriptions and summarize supplied failure logs. Return concise evidence for each category before a developer investigates.
More on GPT-5 Nano and Amp
Overview
GPT-5 Nano in other agents
Questions about GPT-5 Nano in Amp
Can Amp use GPT-5 Nano?
Yes. Add Spiral to Amp as an MCP server, and Amp can find and run GPT-5 Nano like any built-in tool.
How much does GPT-5 Nano cost in Amp?
Up to $0.002134 per run at the default inputs. $0.065 per 1M input tokens, $0.0065 per 1M cached input tokens, $0.52 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Amp plan is separate.
Do I need an OpenAI account to use GPT-5 Nano in Amp?
No. Spiral runs GPT-5 Nano on its own OpenAI account. You sign in to Spiral once and never paste a key.
How long does it take to set up GPT-5 Nano in Amp?
About two minutes: add the server, then sign in, then check it.
Does Amp wait while GPT-5 Nano runs?
GPT-5 Nano runs in the background. Amp starts the run, checks on it and gives you the text, json when it's ready.
What is GPT-5 Nano good at?
OpenAI recommends GPT-5 Nano for summarization and classification. Practical starting points include support ticket labels, short meeting recaps, and extracting named details from text.
How much text can GPT-5 Nano handle?
GPT-5 Nano has a 400,000-token context window and a maximum output allowance of 128,000 tokens. Internal reasoning also uses the output allowance, so visible answers can be shorter.
Does GPT-5 Nano know about recent events?
Its knowledge cutoff is May 31, 2024. For newer information, provide the relevant source text and ask it to answer from that material.
Can GPT-5 Nano return structured JSON?
Yes. Structured Outputs can enforce a supported JSON schema for completed responses that are not refusals, but you still need to check whether the extracted values are correct.
When should I use minimal reasoning?
Start with minimal reasoning for simple labels, extraction, formatting, and short rewrites. OpenAI advises against minimal reasoning for planning with multiple steps or workflows that depend heavily on tools.
Start using GPT-5 Nano in Amp
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Amp:
set up https://spiral.ai/SKILL.md


