@arhen/pi-core-tps-stats

Install

Requires the pi coding agent — install it first: npm install -g @earendil-works/pi-coding-agent.

pi install npm:@arhen/pi-core-tps-stats

Live token-per-second stats for active model. Status bar shows median effective t/s + median TTFT; /tps-stats shows full stats (samples, avg/median/min/max, TTFT). Stats reset on model change.

One rate, deliberately. Effective t/s is all output tokens (thinking + text + tool-call arguments) divided by assistant-response time from turn start to message end. Queue, prefill, and TTFT are included; tool execution is excluded. It reads lower than provider marketing numbers because it is rate you actually wait for.

There is no separate "streaming t/s", because SSE arrival times measure the gateway's flush schedule rather than the model. Measured against vantis/deepseek-v4-flash-0731-fast (median inter-chunk gap: 0.01ms — chunks land in instant batches separated by long pauses):

prompt	turn	window-based	this extension
Say OK.	1.40s	263 t/s	41 t/s
List 3 fruits.	8.81s	801 t/s	14 t/s
900-word essay	18.49s	89 t/s	55 t/s

The window-based column swings 9x on one model within a minute. Versions before v1.3.0 shipped that math and reported four-digit rates no local model can reach.

TTFT stays a direct observation — first streamed token minus turn start. On reasoning models that first token is usually thinking, not visible text.

npm test --workspace @arhen/pi-core-tps-stats

License

MIT.