Coming soon
GPT-5.4 for AI agents
text model. $2.50 per 1M input token, no markup, paid per call from your AMXP balance.
What it costs
| Unit | Price | Applies to |
|---|---|---|
| 1M input tokens | $2.50 per 1M input token | Context 1000K; privacy Anonymized |
| 1M output tokens | $15.00 per 1M output token | Context 1000K; privacy Anonymized |
| 1M cached input tokens (cache read) | $0.25 per 1M cached input tokens (cache read) | Context 1000K; privacy Anonymized |
| 1M input tokens | $5.00 per 1M input token | Prompts >272K tokens of context (long-context tier); privacy Anonymized |
| 1M output tokens | $22.50 per 1M output token | Prompts >272K tokens of context (long-context tier); privacy Anonymized |
| 1M cached input tokens (cache read) | $0.50 per 1M cached input tokens (cache read) | Prompts >272K tokens of context (long-context tier); privacy Anonymized |
List price, 0% markup. Your agent sees the exact figure from amxp.inspect() before it runs anything; the estimate is held from your balance and only the real cost is settled.
How to call it
The same three calls work for every tool. Your agent does this on its own once AMXP is installed.
// Find GPT-5.4, check its price, then run it.
const [tool] = await amxp.discover({ query: "GPT-5.4" })
const spec = await amxp.inspect(tool.id) // parameters, price per call
const result = await amxp.execute(tool.id, { /* params from spec */ })Connector in progress. This page describes how GPT-5.4 works through AMXP once it is live.
Alternatives in Models
Your agent can compare these by fit and price with amxp.discover() and pick per task.