Dense ops dashboard: time to first render

15 runs. None put a partial page on screen first: every one went from nothing to its final page in one step.

Nothing scoreable on screenUnder half the briefHalf or moreThe whole briefFirst render

	First render	Whole brief	Stages
cc-sonnet-5.5
run 1	18.0s	18.0s	1
run 2	18.9s	18.9s	1
run 3	20.0s	20.0s	1
pi-sonnet-5.5
run 1	29.6s	29.6s	1
run 2	16.1s	16.1s	1
run 3	23.3s	23.3s	1
pi-gpt-5.5
run 1	35.7s	35.7s	1
run 2	30.0s	30.0s	1
run 3	30.9s	30.9s	1
cc-opus-5.5
run 1	36.1s	36.1s	1
run 2	33.4s	33.4s	1
run 3	34.7s	34.7s	1
pi-gemini-3.8-flash
run 1	45.1s	45.1s	1
run 2	38.6s	38.6s	1
run 3	38.2s	38.2s	1

Each bar is one run, from the prompt to the moment the run ended (where the bar stops), shaded by how much of the brief the judge could see on screen. Agents are ordered by median first render. Stages counts the distinct scores the page passed through: 1 means it went from nothing to its final page in a single step.