@djoeyg/pi-tps

Output tokens per second on the working spinner and footer status

Packages

Package details

extension

Install @djoeyg/pi-tps from npm and Pi will load the resources declared by the package manifest.

npm repohome report
$ pi install npm:@djoeyg/pi-tps
Package
@djoeyg/pi-tps
Version
0.1.0
Published
Oct 10, 2026
Downloads
183/mo · 183/wk
Author
djoeyg
License
MIT
Types
extension
Size
4.7 KB
Dependencies
0 dependencies · 1 peer
Pi manifest JSON
{
  "extensions": [
    "./index.ts"
  ]
}

Security note

Pi packages can execute code and influence agent behavior. Review the source before installing third-party packages.

README

pi-tps

Shows output tokens per second for assistant responses.

  • While streaming: the working row reads ⠋ Working - 42.3 T/s, estimated from streamed characters (~4 chars/token) and refreshed every 250ms.
  • When a message ends: the rate is recomputed from the provider-reported usage.output and kept in the footer status.

The clock starts at the first streamed content event, so time-to-first-token is excluded and the figure reflects decode speed. Models that reason silently (no thinking deltas streamed) will read high, since their reasoning tokens count toward usage.output but the clock starts late.

Install

pi install npm:@djoeyg/pi-tps