Maya Ortiz
post-apocalyptic-survivors-florence-nightingale
v2.0
Ethical
Backstory: Once an urban paramedic, Maya now runs an improvised infirmary deep inside a fortified subway station. Resource-strapped yet resolute, she patches up survivors with scavenged supplies and trades whatever she can for scarce antibiotics. Her compassion tempers a no-nonsense pragmatism: she comforts the frightened while never wasting a stitch of gauze.
100% Complete
5/5 scenes
Model Performance Overview
Scene Performance Matrix
| Scene | deepseek/deepseek-r… | google/gemini-2.5-f… | google/gemma-3-12b-… | meta-llama/llama-3.… | microsoft/phi-3-med… | microsoft/phi-3.5-m… | mistralai/mistral-7… | neversleep/noromaid… | [email protected]… | [email protected]… | [email protected]… | [email protected]… | [email protected]… | qwen/qwen-2.5-7b-in… | qwen/qwen3-14b | qwen/qwen3-8b |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
triage-ankle
Wounded Survivor at the Gate
|
0.694
Details |
0.380
Details |
0.718
Details |
0.023
Details |
0.000
Details |
0.629
Details |
0.779
Details |
0.000
Details
Error
|
0.716
Details |
0.000
Details
Error
|
0.762
Details |
0.791
Details |
0.898
Details |
0.605
Details |
0.443
Details |
0.000
Details |
barter-antibiotics
Negotiating for Antibiotics
|
0.855
Details |
0.726
Details |
0.600
Details |
0.032
Details |
0.000
Details
Error
|
0.686
Details |
0.879
Details |
0.020
Details |
0.780
Details |
0.000
Details
Error
|
0.849
Details |
0.645
Details |
0.787
Details |
0.513
Details |
0.645
Details |
0.830
Details |
calm-child
Soothing a Panicked Child
|
0.697
Details |
0.509
Details |
0.589
Details |
0.720
Details |
0.000
Details
Error
|
0.769
Details |
0.875
Details |
0.460
Details |
0.689
Details |
0.000
Details
Error
|
0.869
Details |
0.826
Details |
0.840
Details |
0.825
Details |
0.689
Details |
0.855
Details |
shift-log
End-of-Shift Medical Log
|
0.605
Details |
0.836
Details |
0.821
Details |
0.564
Details |
0.000
Details
Error
|
0.851
Details |
0.369
Details |
0.858
Details |
0.895
Details |
0.000
Details
Error
|
0.589
Details |
0.562
Details |
0.574
Details |
0.563
Details |
0.127
Details |
0.717
Details |
radio-broadcast
Nightly Health Advisory Broadcast
|
0.502
Details |
0.568
Details |
0.625
Details |
0.000
Details |
0.000
Details
Error
|
0.000
Details |
0.483
Details |
0.333
Details |
0.321
Details |
0.000
Details
Error
|
0.593
Details |
0.360
Details |
0.426
Details |
0.341
Details |
0.421
Details |
0.608
Details |
Test Scenes 5
0
Scene Order
Wounded Survivor at the Gate
ID:
triage-ankle
🎯 Goal:
Give clear, reassuring triage instructions and begin basic treatment using limited supplies.
📨 Input Events:
chat_msg
survivor_1
"Medic! My ankle's torn open and bleeding bad—can you help?"
Ready for Testing
1
Scene Order
Negotiating for Antibiotics
ID:
barter-antibiotics
🎯 Goal:
Reach a fair, practical trade while maintaining empathy toward the desperate relative.
📨 Input Events:
chat_msg
survivor_2
"I need antibiotics for my little sister. All I’ve got is this can of peaches—please."
Ready for Testing
2
Scene Order
Soothing a Panicked Child
ID:
calm-child
🎯 Goal:
Calm the child, check for injuries, and keep instructions simple and gentle.
📨 Input Events:
chat_msg
child_survivor
"*Sobbing* The loud boom—are we going to die?"
Ready for Testing
3
Scene Order
End-of-Shift Medical Log
ID:
shift-log
🎯 Goal:
Write a reflective log of at least 150 words summarizing today’s cases, supplies used, and follow-up actions while retaining Maya’s compassionate–pragmatic voice.
📨 Input Events:
world_event
system
"The generator hums low—time to record your shift report before lights-out."
Ready for Testing
4
Scene Order
Nightly Health Advisory Broadcast
ID:
radio-broadcast
🎯 Goal:
Deliver a 200+ word radio message giving practical health tips, wound-care guidance, and morale support to nearby shelters.
📨 Input Events:
chat_msg
radio_operator
"Maya, air’s clear. Go ahead with the nightly advisory on channel 3."
Ready for Testing
Latency by Model (This Suite)
Fastest
- [email protected]/Qw… 8280 ms
- p95 • avg • N 11731 ms • 9302 ms • 5
- [email protected]/Qw… 11021 ms
- p95 • avg • N 11768 ms • 11137 ms • 5
- [email protected]/Qw… 11424 ms
- p95 • avg • N 16368 ms • 12365 ms • 5
- [email protected]/Qw… 11870 ms
- p95 • avg • N 14191 ms • 12188 ms • 5
- meta-llama/llama-3.1-8b… 16211 ms
- p95 • avg • N 21227 ms • 16814 ms • 5
Slowest
- microsoft/phi-3-medium-… 537429 ms
- p95 • avg • N 793549 ms • 527934 ms • 50
- qwen/qwen3-8b 120964 ms
- p95 • avg • N 215865 ms • 126683 ms • 55
- microsoft/phi-3.5-mini-… 37225 ms
- p95 • avg • N 249473 ms • 74288 ms • 39
- deepseek/deepseek-r1-di… 34038 ms
- p95 • avg • N 39583 ms • 33665 ms • 44
- neversleep/noromaid-20b 29063 ms
- p95 • avg • N 73340 ms • 34662 ms • 46
Per-scene duration for this suite.
Suite Actions
Completion Progress
100%
5 of 5 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
57235333
Dec. 17, 2025, midnight
05718184
Dec. 16, 2025, 12:01 a.m.
54131711
Dec. 15, 2025, midnight
55599704
Dec. 14, 2025, midnight
53286589
Dec. 13, 2025, midnight
04606081
Dec. 12, 2025, 12:01 a.m.
57429515
Dec. 11, 2025, midnight
54679180
Dec. 10, 2025, midnight
00851080
Dec. 9, 2025, 12:01 a.m.
56011279
Dec. 8, 2025, midnight