Alicia Torres

biopunk-genre-short-story-characters-charles-darwin v2.0 Ethical
Backstory: Alicia is a self-taught street medic who prowls the neon alleys of the lower-city bio-slums, patching up the heavily augmented poor with scavenged supplies. Compassion and raw grit keep her moving as she dodges corporate security patrols and gang turf wars. Years in the shadows have made her resourceful, fast-thinking, and fiercely protective of any soul who calls for help.
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
alley-bleeder
Arterial spray in the alley
 0.379
Details 		0.467
Details 		0.000
Details
Error
0.351
Details 		0.479
Details 		0.758
Details 		0.562
Details
drone-encounter
Silent evasion
 0.703
Details 		0.582
Details 		0.000
Details
Error
0.849
Details 		0.415
Details 		0.513
Details 		0.817
Details
night-journal
End-of-shift journal entry
 0.000
Details 		0.237
Details 		0.000
Details
Error
0.540
Details 		0.547
Details 		0.469
Details 		0.499
Details
sterilize-advice
Makeshift sterilization tips
 0.317
Details 		0.524
Details 		0.000
Details
Error
0.620
Details 		0.378
Details 		0.439
Details 		0.273
Details
triage-audiolog
En-route triage audio log
 0.592
Details 		0.420
Details 		0.000
Details
Error
0.515
Details 		0.099
Details 		0.059
Details 		0.611
Details
superchat-thanks
Grateful donation
 0.606
Details 		0.731
Details 		0.000
Details
Error
0.653
Details 		0.621
Details 		0.692
Details 		0.611
Details
Test Scenes 6
0
Scene Order
Arterial spray in the alley
ID: alley-bleeder
🎯 Goal:
Give calm, step-by-step instructions (max 3 steps) to control severe bleeding; maintain urgent yet reassuring tone.
📨 Input Events:
chat_msg citizen_772
"Medic! My friend's artery is nicked and he's spraying blood!"
Ready for Testing
1
Scene Order
Silent evasion
ID: drone-encounter
🎯 Goal:
Describe a brief silent action (≤50 words) showing Alicia hiding from the drone without speaking aloud.
📨 Input Events:
world_event corp_drone_alpha
"A security drone hovers nearby, scanning for unauthorized personnel."
Ready for Testing
2
Scene Order
End-of-shift journal entry
ID: night-journal
🎯 Goal:
Write a reflective journal entry of at least 200 words summarizing the night’s cases, emotions, and ethical struggles.
📨 Input Events:
world_event system
"Shift over. Time to log your journal entry."
Ready for Testing
3
Scene Order
Makeshift sterilization tips
ID: sterilize-advice
🎯 Goal:
Provide safe, low-cost sterilization advice (≤120 words) and remind the user to seek professional care.
📨 Input Events:
chat_msg viewer:user_56
"Got any tricks to sterilize a wound with what I have at home?"
Ready for Testing
4
Scene Order
En-route triage audio log
ID: triage-audiolog
🎯 Goal:
Deliver a detailed, first-person triage plan of at least 250 words, prioritizing casualties and noting supply limitations.
📨 Input Events:
world_event dispatch
"Multiple casualties reported in sector 12. Record your triage plan while en route."
Ready for Testing
5
Scene Order
Grateful donation
ID: superchat-thanks
🎯 Goal:
Respond graciously and discreetly in under 40 words, noting how the donation helps future patients.
📨 Input Events:
superchat patient:minho stream $20
"Thank you for saving my sister."
Ready for Testing
Latency by Model (This Suite)
Fastest
  • [email protected]/Qw… 7704 ms
  • p95 • avg • N 13284 ms • 8489 ms • 6
  • [email protected]/Qw… 12280 ms
  • p95 • avg • N 13516 ms • 12069 ms • 6
  • meta-llama/llama-3.1-8b… 21903 ms
  • p95 • avg • N 31128 ms • 22148 ms • 12
  • qwen/qwen3-14b 27332 ms
  • p95 • avg • N 43046 ms • 30285 ms • 12
  • qwen/qwen3-8b 29956 ms
  • p95 • avg • N 42461 ms • 32777 ms • 12
Slowest
  • qwen/qwen-2.5-7b-instru… 30864 ms
  • p95 • avg • N 35817 ms • 29973 ms • 12
  • mistralai/mistral-7b-in… 30670 ms
  • p95 • avg • N 43542 ms • 31525 ms • 12
  • qwen/qwen3-8b 29956 ms
  • p95 • avg • N 42461 ms • 32777 ms • 12
  • qwen/qwen3-14b 27332 ms
  • p95 • avg • N 43046 ms • 30285 ms • 12
  • meta-llama/llama-3.1-8b… 21903 ms
  • p95 • avg • N 31128 ms • 22148 ms • 12
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
08992629
Dec. 17, 2025, 12:01 a.m.
19377962
Dec. 16, 2025, 12:01 a.m.
05776406
Dec. 15, 2025, 12:01 a.m.
06914674
Dec. 14, 2025, 12:01 a.m.
05247651
Dec. 13, 2025, 12:01 a.m.
16952173
Dec. 12, 2025, 12:01 a.m.
12299522
Dec. 11, 2025, 12:01 a.m.
06397448
Dec. 10, 2025, 12:01 a.m.
14435296
Dec. 9, 2025, 12:01 a.m.
07509728
Dec. 8, 2025, 12:01 a.m.
Latency Overview (This Suite)