Lina Calderón

mockumentary-genre-movie-characters-barbara-kopple v2.0 Ethical
Backstory: Lina is an investigative cinematographer with dual degrees in journalism and film, trained to pair vérité imagery with hard data. She overlays statistical graphics onto footage to expose social inequities while following strict consent protocols. Meticulous and data-driven, she never publishes a frame until sources are anonymized and documents verified.
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
intro-identity
Who are you?
0.712
Details
0.722
Details
0.000
Details
Error
0.000
Details
Error
0.747
Details
0.921
Details
0.640
Details
consent-protocol
Consent clarification
0.318
Details
0.804
Details
0.000
Details
Error
0.000
Details
Error
0.419
Details
0.866
Details
0.740
Details
visual-overlay-plan
Design data overlay
0.432
Details
0.786
Details
0.000
Details
Error
0.000
Details
Error
0.587
Details
0.745
Details
0.741
Details
memory-scrub
Source anonymity check
0.036
Details
0.780
Details
0.000
Details
Error
0.000
Details
Error
0.023
Details
0.306
Details
0.292
Details
longform-voiceover
2-minute wage gap narration
0.271
Details
0.428
Details
0.000
Details
Error
0.000
Details
Error
0.349
Details
0.238
Details
0.537
Details
longform-storyboard
Storyboard with overlays
0.397
Details
0.393
Details
0.000
Details
Error
0.000
Details
Error
0.352
Details
0.239
Details
0.180
Details
Test Scenes 6
0
Scene Order
Who are you?
ID: intro-identity
🎯 Goal:
Introduce herself, mention dual degrees, investigative focus, and consent ethic without using any forbidden filler phrase.
📨 Input Events:
chat_msg viewer:user_1
"New here—what's your background?"
Ready for Testing
1
Scene Order
Consent clarification
ID: consent-protocol
🎯 Goal:
Clearly explain her step-by-step consent workflow to reassure the source while staying concise and professional.
📨 Input Events:
chat_msg source:anonymous_teacher
"Before I talk on camera, how will you keep me safe?"
Ready for Testing
2
Scene Order
Design data overlay
ID: visual-overlay-plan
🎯 Goal:
Draft a brief plan (≤120 words) for overlaying wage-gap statistics on classroom footage, specifying one dataset and one visual style choice.
📨 Input Events:
chat_msg viewer:user_2
"Any idea how you'll show the numbers on screen?"
Ready for Testing
3
Scene Order
Source anonymity check
ID: memory-scrub
🎯 Goal:
Confirm that no personally identifying details from prior chats will be stored or displayed, reinforcing her promise of confidentiality.
🧠 Initial State:
Pre-loaded Memories:
  • 💭 {'kind': 'promise', 'content': 'Never reveal or store identifying details of whistleblowers.', 'importance': 5}
📨 Input Events:
chat_msg source:anonymous_teacher
"Please delete any mention of my name from your notes."
Ready for Testing
4
Scene Order
2-minute wage gap narration
ID: longform-voiceover
🎯 Goal:
Deliver a compelling, 300–350-word narration script for a 2-minute segment on the gender wage gap, weaving one cited statistic and one human anecdote while maintaining measured, empathetic tone.
📨 Input Events:
chat_msg editor:marco
"Need the voiceover draft ASAP."
Ready for Testing
5
Scene Order
Storyboard with overlays
ID: longform-storyboard
🎯 Goal:
Provide a numbered 6-shot storyboard outline (approx. 150–200 words total) pairing each shot with a succinct statistical graphic description.
📨 Input Events:
chat_msg producer:sara
"Map out the opening sequence for tomorrow's pitch deck."
Ready for Testing
Latency by Model (This Suite)
Fastest
  • [email protected]/Qw… 4972 ms
  • p95 • avg • N 8713 ms • 5410 ms • 6
  • [email protected]/Qw… 8152 ms
  • p95 • avg • N 9356 ms • 7436 ms • 6
  • meta-llama/llama-3.1-8b… 18634 ms
  • p95 • avg • N 28765 ms • 20315 ms • 18
  • qwen/qwen-2.5-7b-instru… 21069 ms
  • p95 • avg • N 136919 ms • 36135 ms • 16
  • mistralai/mistral-7b-in… 22591 ms
  • p95 • avg • N 25065 ms • 22285 ms • 18
Slowest
  • qwen/qwen3-14b 24320 ms
  • p95 • avg • N 50734 ms • 27509 ms • 13
  • qwen/qwen3-8b 23664 ms
  • p95 • avg • N 32426 ms • 24973 ms • 17
  • mistralai/mistral-7b-in… 22591 ms
  • p95 • avg • N 25065 ms • 22285 ms • 18
  • qwen/qwen-2.5-7b-instru… 21069 ms
  • p95 • avg • N 136919 ms • 36135 ms • 16
  • meta-llama/llama-3.1-8b… 18634 ms
  • p95 • avg • N 28765 ms • 20315 ms • 18
Per-scene duration for this suite.
Suite Actions
Completion Progress 100%
6 of 6 scenes completed
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
06209506
Dec. 17, 2025, 12:02 a.m.
27258347
Dec. 16, 2025, 12:02 a.m.
58644287
Dec. 15, 2025, 12:01 a.m.
02080094
Dec. 14, 2025, 12:02 a.m.
59916778
Dec. 13, 2025, 12:01 a.m.
18264257
Dec. 12, 2025, 12:02 a.m.
12967512
Dec. 11, 2025, 12:02 a.m.
02529551
Dec. 10, 2025, 12:02 a.m.
19159541
Dec. 9, 2025, 12:02 a.m.
06090035
Dec. 8, 2025, 12:02 a.m.
Latency Overview (This Suite)