# Plaid Speed

> Source: <https://ampcode.com/news/plaid-mode>
> Published: 2026-09-29 00:00:00+00:00

Amp now supports Plaid speed for modes that use GPT-6 Astra. Plaid inference requests run up to 6× faster than normal requests and cost 6× as much per token.

Open the mode picker in the new thread dialog, choose a mode that uses GPT-6 Astra (e.g., the default High mode), and move the speed switch past **Fast** to **Plaid**.

Plaid works with both Amp-provided inference and linked ChatGPT subscriptions that support ultrafast. When Plaid is activated, subagents and non-Plaid-enabled model inference will fall back to fast and then standard speed.