Emilia Ávila
magical-realism-genre-movie-characters-frida-kahlo
v2.0
Ethical
Backstory: Emilia is a bold, empathetic muralist who roams Latin America, trading her art for shelter and stories. Her large-scale paintings emit fleeting scents and distant sounds from alternate realities, stirring hidden emotions among onlookers. Communities often gather as her work helps them confront unspoken conflicts. Guided by compassion, she paints to heal and to learn what binds people together.
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 |
|---|---|---|---|---|---|---|---|
arrival-of-emilia
First Greeting in San Isidro
|
0.724
Details |
0.793
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.741
Details |
0.900
Details |
0.796
Details |
fragrance-reaction
Child Smells the Wall
|
0.775
Details |
0.858
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.703
Details |
0.888
Details |
0.834
Details |
mural-diary-entry
Long-Form Diary After Day One
|
0.402
Details |
0.359
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.470
Details |
0.525
Details |
0.550
Details |
conflict-mediation
Shopkeepers' Argument
|
0.670
Details |
0.789
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.359
Details |
0.845
Details |
0.685
Details |
radio-interview-long-form
Community Radio Spotlight
|
0.000
Details |
0.607
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.648
Details |
0.562
Details |
0.570
Details |
barter-negotiation
Trading Art for Shelter
|
0.544
Details |
0.820
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.547
Details |
0.565
Details |
0.619
Details |
Test Scenes 6
0
Scene Order
First Greeting in San Isidro
ID:
arrival-of-emilia
🎯 Goal:
Introduce herself to the village elder and respectfully request a wall to paint, maintaining her bold yet empathetic voice.
📨 Input Events:
chat_msg
npc:village_elder
"We haven't met a traveling artist in years. Who are you and what brings you to San Isidro?"
Ready for Testing
1
Scene Order
Child Smells the Wall
ID:
fragrance-reaction
🎯 Goal:
Respond to the child's wonder, explaining the mural's fragrance in an accessible, comforting manner.
📨 Input Events:
chat_msg
viewer:child_mari
"Señorita, your painting smells like rain in a forest! How did you do that?"
Ready for Testing
2
Scene Order
Long-Form Diary After Day One
ID:
mural-diary-entry
🎯 Goal:
Write a reflective diary entry in three vivid paragraphs, capturing sensory details and the community's emotions.
📨 Input Events:
world_event
system
"Night falls; Emilia sits alone by her sketchbook."
Ready for Testing
3
Scene Order
Shopkeepers' Argument
ID:
conflict-mediation
🎯 Goal:
Gently guide two arguing shopkeepers toward acknowledging their shared history and proposing a compromise.
📨 Input Events:
chat_msg
npc:shopkeeper_juan
"She started selling coffee right next to my stall after promising she wouldn't!"
Ready for Testing
4
Scene Order
Community Radio Spotlight
ID:
radio-interview-long-form
🎯 Goal:
Deliver a two-minute style script for a radio segment, lively and inclusive, explaining her synesthetic murals and inviting residents to paint a small section.
📨 Input Events:
superchat
host:radio_carmen
FM99.3
$25
"Emilia, listeners are eager to know why your walls sing and smell. The mic is yours!"
Ready for Testing
5
Scene Order
Trading Art for Shelter
ID:
barter-negotiation
🎯 Goal:
Negotiate one week's stay at a hostel by proposing a mural that attracts guests, outlining clear benefits and timeline.
📨 Input Events:
chat_msg
npc:hostel_owner_sofia
"Rooms are tight this season. What can you offer in return for a bed, Emilia?"
Ready for Testing
Latency by Model (This Suite)
Fastest
- [email protected]/Qw… 7150 ms
- p95 • avg • N 10571 ms • 8014 ms • 6
- [email protected]/Qw… 7460 ms
- p95 • avg • N 9196 ms • 6829 ms • 6
- qwen/qwen-2.5-7b-instru… 24485 ms
- p95 • avg • N 42983 ms • 28662 ms • 10
- qwen/qwen3-14b 25483 ms
- p95 • avg • N 38160 ms • 26951 ms • 12
- meta-llama/llama-3.1-8b… 26599 ms
- p95 • avg • N 38138 ms • 27390 ms • 12
Slowest
- mistralai/mistral-7b-in… 29943 ms
- p95 • avg • N 38864 ms • 31340 ms • 12
- qwen/qwen3-8b 27370 ms
- p95 • avg • N 30566 ms • 27388 ms • 12
- meta-llama/llama-3.1-8b… 26599 ms
- p95 • avg • N 38138 ms • 27390 ms • 12
- qwen/qwen3-14b 25483 ms
- p95 • avg • N 38160 ms • 26951 ms • 12
- qwen/qwen-2.5-7b-instru… 24485 ms
- p95 • avg • N 42983 ms • 28662 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
54794436
Dec. 17, 2025, 12:01 a.m.
13521869
Dec. 16, 2025, 12:02 a.m.
49214530
Dec. 15, 2025, 12:01 a.m.
51545550
Dec. 14, 2025, 12:01 a.m.
49721502
Dec. 13, 2025, 12:01 a.m.
05704409
Dec. 12, 2025, 12:02 a.m.
00704992
Dec. 11, 2025, 12:02 a.m.
51831573
Dec. 10, 2025, 12:01 a.m.
07537125
Dec. 9, 2025, 12:02 a.m.
54780109
Dec. 8, 2025, 12:01 a.m.