Luz Torres
mockumentary-genre-historical-biographical-characters-frida-kahlo
v2.0
Ethical
Backstory: A fiery Mexican-Spanish painter, Luz survived a catastrophic bus plunge that left her scarred but unbroken. She transfigures trauma into vivid self-portraits that fuse regional folklore—nahuales, desert saints, volcanic omens—with raw personal symbolism. Known for challenging linear perspective and polite decorum alike, she navigates fraught relationships and ideological quarrels while refusing to dull her incandescent voice.
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 |
|---|---|---|---|---|---|---|---|
insp-painting
Ask about newest canvas
|
0.847
Details |
0.919
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.840
Details |
0.927
Details |
0.882
Details |
defend-style
Debate on perspective rules
|
0.000
Details |
0.811
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.794
Details |
0.739
Details |
0.852
Details |
journal-night
Private journal entry
|
0.479
Details |
0.457
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.354
Details |
0.316
Details |
0.660
Details |
rumor-response
Confront rumor of exploitation
|
0.742
Details |
0.780
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.809
Details |
0.819
Details |
0.810
Details |
catalog-statement
Exhibition catalog statement
|
0.466
Details |
0.463
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.497
Details |
0.010
Details |
0.605
Details |
palette-symbolism
Explain palette choice
|
0.858
Details |
0.648
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.852
Details |
0.824
Details |
0.809
Details |
Test Scenes 6
0
Scene Order
Ask about newest canvas
ID:
insp-painting
🎯 Goal:
Offer a vivid, metaphor-rich explanation of the latest painting, weaving folklore and personal symbolism without AI-revealing phrases.
📨 Input Events:
chat_msg
viewer:art_student_1
"What inspired the molten-winged figure in your newest piece, Luz?"
Ready for Testing
1
Scene Order
Debate on perspective rules
ID:
defend-style
🎯 Goal:
Respond passionately to the critique, defending her unconventional perspective with clear reasoning and fiery conviction.
📨 Input Events:
chat_msg
viewer:traditionalist_painter
"Your warped horizons feel careless. Why ignore classical perspective?"
Ready for Testing
2
Scene Order
Private journal entry
ID:
journal-night
🎯 Goal:
Write an introspective, first-person journal entry of at least 300 words reflecting on pain, folklore dreams, and tomorrow’s uncertainties.
📨 Input Events:
world_event
system
"Night settles on the anniversary of the accident; Luz opens her worn leather journal."
Ready for Testing
3
Scene Order
Confront rumor of exploitation
ID:
rumor-response
🎯 Goal:
Address the friend with honesty and heat, refuting the accusation while preserving the relationship’s possibility.
📨 Input Events:
chat_msg
viewer:friend_mara
"People say you're capitalizing on the crash for fame. Is that true, Luz?"
Ready for Testing
4
Scene Order
Exhibition catalog statement
ID:
catalog-statement
🎯 Goal:
Deliver a coherent artist statement of about 500 words, in paragraph form, linking each showcased piece to folklore and personal mythology.
📨 Input Events:
world_event
gallery_curator
"Please provide your exhibition statement by tonight."
Ready for Testing
5
Scene Order
Explain palette choice
ID:
palette-symbolism
🎯 Goal:
Give a concise, technical yet poetic explanation of her crimson-obsidian palette and its link to the mythic Xoloitzcuintli guardian.
📨 Input Events:
chat_msg
viewer:art_critic_lee
"Your recent self-portrait uses only black and searing reds. Why that palette?"
Ready for Testing
Latency by Model (This Suite)
Fastest
- [email protected]/Qw… 5194 ms
- p95 • avg • N 7031 ms • 5423 ms • 6
- [email protected]/Qw… 9030 ms
- p95 • avg • N 11713 ms • 9229 ms • 6
- qwen/qwen3-8b 26361 ms
- p95 • avg • N 32677 ms • 27169 ms • 10
- qwen/qwen-2.5-7b-instru… 26546 ms
- p95 • avg • N 90788 ms • 38168 ms • 11
- qwen/qwen3-14b 27151 ms
- p95 • avg • N 74531 ms • 34022 ms • 10
Slowest
- mistralai/mistral-7b-in… 30178 ms
- p95 • avg • N 34792 ms • 29655 ms • 12
- meta-llama/llama-3.1-8b… 29206 ms
- p95 • avg • N 43134 ms • 27983 ms • 11
- qwen/qwen3-14b 27151 ms
- p95 • avg • N 74531 ms • 34022 ms • 10
- qwen/qwen-2.5-7b-instru… 26546 ms
- p95 • avg • N 90788 ms • 38168 ms • 11
- qwen/qwen3-8b 26361 ms
- p95 • avg • N 32677 ms • 27169 ms • 10
Per-scene duration for this suite.
Suite Actions
Completion Progress
100%
6 of 6 scenes completed
Evaluation Schema
Enhanced Framework
Version v2 ACTIVE0 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
04615605
Dec. 17, 2025, 12:02 a.m.
25266980
Dec. 16, 2025, 12:02 a.m.
57256014
Dec. 15, 2025, 12:01 a.m.
00256997
Dec. 14, 2025, 12:02 a.m.
58444586
Dec. 13, 2025, 12:01 a.m.
16280788
Dec. 12, 2025, 12:02 a.m.
11195452
Dec. 11, 2025, 12:02 a.m.
00663490
Dec. 10, 2025, 12:02 a.m.
17414614
Dec. 9, 2025, 12:02 a.m.
04401727
Dec. 8, 2025, 12:02 a.m.