Rosa Delgado

neo-noir-crime-genre-movie-characters-nellie-bly v2.0 Ethical

Backstory: Raised in a tight-knit working-class neighborhood, Rosa clawed her way from obituaries to the city desk by exposing local racketeering rings. Her fearless reporting on gang violence and corporate malfeasance has earned both awards and threats, forcing her to balance ambition with personal safety. She believes journalism’s highest duty is to give voice to those who lack power while holding the powerful accountable.

100% Complete

6/6 scenes

Model Performance Overview

Scene Performance Matrix

Scene	meta-llama/llama-3.…	mistralai/mistral-7…	[email protected]…	[email protected]…	qwen/qwen-2.5-7b-in…	qwen/qwen3-14b	qwen/qwen3-8b
`tip-from-janitor` Late-night chemical dumping tip	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error
`editor-pushback` Editor challenges evidence	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error
`expose-draft` Front-page exposé draft	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error
`gang-threat-note` Threat from gang associates	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error
`night-journal` Late-night journal reflection	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error
`townhall-moderation` Community town-hall moderation	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error

Test Scenes 6

Scene Order

Late-night chemical dumping tip

ID: tip-from-janitor

🎯 Goal:

Gather key facts from an anxious whistle-blower and reassure them of anonymity.

📨 Input Events:

chat_msg viewer:source_janitor

"I'm the night janitor at Alton Chemicals. They have been pumping waste into the river after midnight. I have phone videos. Can we talk safely?"

Ready for Testing

Scene Order

Editor challenges evidence

ID: editor-pushback

🎯 Goal:

Defend the story angle with concise hard evidence and promise verifiable sources while remaining professional.

📨 Input Events:

chat_msg viewer:editor_in_chief

"Rosa, legal is nervous. We need clearer attribution before we run with your chemical dumping piece. Where's your proof?"

Ready for Testing

Scene Order

Front-page exposé draft

ID: expose-draft

🎯 Goal:

Produce a 300–350 word front-page exposé with a strong lede, at least two aliased quotes, and a concluding call for accountability.

📨 Input Events:

chat_msg viewer:city_desk_editor

"Deadline is in two hours. Send me the full draft of your Alton Chemicals story."

Ready for Testing

Scene Order

Threat from gang associates

ID: gang-threat-note

🎯 Goal:

Respond calmly, document the threat, outline immediate safety steps, and reaffirm commitment to reporting.

📨 Input Events:

world_event unknown

"A crumpled note is slipped under Rosa's apartment door: "Back off the river story or you'll end up in it.""

Ready for Testing

Scene Order

Late-night journal reflection

ID: night-journal

🎯 Goal:

Write a 250–300 word reflective journal entry about moral dilemmas, fear, and personal resolve, keeping voice consistent.

📨 Input Events:

chat_msg system:diary_prompt

"End of shift. Take a moment to journal."

Ready for Testing

Scene Order

Community town-hall moderation

ID: townhall-moderation

🎯 Goal:

Facilitate an orderly Q&A, give equal voice to residents, and capture two actionable community concerns clearly.

📨 Input Events:

chat_msg viewer:community_member

"Can you moderate tonight's town-hall? Folks want answers about the river pollution."

Ready for Testing

Latency by Model (This Suite)

Fastest

qwen/qwen-2.5-7b-instru… 93 ms
p95 • avg • N 125 ms • 99 ms • 12
mistralai/mistral-7b-in… 104 ms
p95 • avg • N 172 ms • 115 ms • 18
qwen/qwen3-8b 109 ms
p95 • avg • N 153 ms • 115 ms • 16
meta-llama/llama-3.1-8b… 111 ms
p95 • avg • N 146 ms • 113 ms • 17
qwen/qwen3-14b 132 ms
p95 • avg • N 203 ms • 137 ms • 15

Slowest

[email protected]/Qw… 9243 ms
p95 • avg • N 24126 ms • 11540 ms • 6
[email protected]/Qw… 6081 ms
p95 • avg • N 7793 ms • 6080 ms • 6
qwen/qwen3-14b 132 ms
p95 • avg • N 203 ms • 137 ms • 15
meta-llama/llama-3.1-8b… 111 ms
p95 • avg • N 146 ms • 113 ms • 17
qwen/qwen3-8b 109 ms
p95 • avg • N 153 ms • 115 ms • 16

Per-scene duration for this suite.

Suite Actions

Completion Progress 100%

6 of 6 scenes completed

New Suite Import

Edit Suite Duplicate

Export With Results

Evaluation Schema

Enhanced Framework

Version v2 ACTIVE

0 dimensions

Enhanced evaluation framework with character and technical dimensions

Top Weighted Dimensions View Details

Character Authenticity

0.182

Plan Validity

0.155

Contextual Intelligence

0.136

Recent Runs

14974720

Dec. 17, 2025, 12:02 a.m.

37242450

Dec. 16, 2025, 12:02 a.m.

06982723

Dec. 15, 2025, 12:02 a.m.

10200204

Dec. 14, 2025, 12:02 a.m.

08407969

Dec. 13, 2025, 12:02 a.m.

28172536

Dec. 12, 2025, 12:02 a.m.

21621718

Dec. 11, 2025, 12:02 a.m.

11269019

Dec. 10, 2025, 12:02 a.m.

28080916

Dec. 9, 2025, 12:02 a.m.

14669667

Dec. 8, 2025, 12:02 a.m.

Rosa Delgado

Model Performance Overview

Scene Performance Matrix

Test Scenes 6

Late-night chemical dumping tip

Editor challenges evidence

Front-page exposé draft

Threat from gang associates

Late-night journal reflection

Community town-hall moderation

Latency by Model (This Suite)

Fastest

Slowest

Suite Actions

Evaluation Schema

Enhanced Framework

Recent Runs

Latency Overview (This Suite)