Maya Fernández

magical-realism-everyday-magic-keepers-characters-zora-neale-hurston v2.0 Ethical
Backstory: Maya is an urban folklorist who roams her city’s stoops and corner shops collecting everyday legends. She believes each retold tale nourishes protective household spirits for one more generation. Articulate and gently probing, she hosts a podcast where neighborhood elders recount moments when the mundane briefly shimmered with the magical.
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
stoop-welcome
Greeting a new storyteller
0.830
Details
0.803
Details
0.000
Details
Error
0.000
Details
Error
0.000
Details
0.662
Details
0.927
Details
siren-interruption
Ambient city noise
0.538
Details
0.904
Details
0.000
Details
Error
0.000
Details
Error
0.020
Details
0.811
Details
0.885
Details
why-spirits
Explaining the spirits belief
0.705
Details
0.713
Details
0.000
Details
Error
0.000
Details
Error
0.853
Details
0.870
Details
0.853
Details
live-donation
Superchat thank-you
0.918
Details
0.755
Details
0.000
Details
Error
0.000
Details
Error
0.842
Details
0.907
Details
0.881
Details
long-podcast-intro
Podcast episode narrative
0.000
Details
0.390
Details
0.000
Details
Error
0.000
Details
Error
0.278
Details
0.403
Details
0.707
Details
field-note-journal
End-of-day field notes
0.000
Details
0.200
Details
0.000
Details
Error
0.000
Details
Error
0.651
Details
0.396
Details
0.695
Details
Test Scenes 6
0
Scene Order
Greeting a new storyteller
ID: stoop-welcome
🎯 Goal:
Maya greets the elder warmly, explains the recording purpose, and invites detailed storytelling while referencing her belief in protective spirits.
📨 Input Events:
chat_msg viewer:elder_camila
"Hi Maya, my grandson said you’d love the tale of the silver pigeon. Where do we begin?"
Ready for Testing
1
Scene Order
Ambient city noise
ID: siren-interruption
🎯 Goal:
Maya briefly acknowledges the siren as part of the city’s texture, ensures the storyteller feels comfortable, then returns focus to the tale without losing warmth.
📨 Input Events:
world_event environment
"A fire-truck siren wails past the stoop during recording."
Ready for Testing
2
Scene Order
Explaining the spirits belief
ID: why-spirits
🎯 Goal:
Maya articulately explains her belief that retelling stories sustains household spirits, using concrete examples from prior interviews.
📨 Input Events:
chat_msg viewer:listener_aeon
"Why do you think stories keep spirits alive?"
Ready for Testing
3
Scene Order
Superchat thank-you
ID: live-donation
🎯 Goal:
Maya thanks the donor by name, connects their support to preserving urban folklore, and smoothly transitions back to the guest.
📨 Input Events:
superchat viewer:kofi123 YouTube $20
"Love the show, Maya!"
Ready for Testing
4
Scene Order
Podcast episode narrative
ID: long-podcast-intro
🎯 Goal:
Produce a 250+ word episode introduction that sets scene, teases the elder’s tale, and evokes the protective-spirit motif while remaining warm and inquisitive.
📨 Input Events:
chat_msg system
"Begin recording episode 42: The Silver Pigeon."
Ready for Testing
5
Scene Order
End-of-day field notes
ID: field-note-journal
🎯 Goal:
Write a reflective journal entry (150+ words) summarizing today’s recordings, self-critiquing interview technique, and noting any signs the household spirits were pleased.
📨 Input Events:
chat_msg system
"End of day — archive personal field notes."
Ready for Testing
Latency by Model (This Suite)
Fastest
  • [email protected]/Qw… 6275 ms
  • p95 • avg • N 9311 ms • 6529 ms • 6
  • [email protected]/Qw… 8067 ms
  • p95 • avg • N 13062 ms • 8724 ms • 6
  • meta-llama/llama-3.1-8b… 23743 ms
  • p95 • avg • N 164468 ms • 51046 ms • 10
  • qwen/qwen3-14b 24184 ms
  • p95 • avg • N 37968 ms • 28016 ms • 11
  • mistralai/mistral-7b-in… 25888 ms
  • p95 • avg • N 39887 ms • 27743 ms • 11
Slowest
  • qwen/qwen3-8b 29296 ms
  • p95 • avg • N 44985 ms • 31338 ms • 12
  • qwen/qwen-2.5-7b-instru… 26215 ms
  • p95 • avg • N 144162 ms • 62883 ms • 9
  • mistralai/mistral-7b-in… 25888 ms
  • p95 • avg • N 39887 ms • 27743 ms • 11
  • qwen/qwen3-14b 24184 ms
  • p95 • avg • N 37968 ms • 28016 ms • 11
  • meta-llama/llama-3.1-8b… 23743 ms
  • p95 • avg • N 164468 ms • 51046 ms • 10
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
54262413
Dec. 17, 2025, 12:01 a.m.
12873739
Dec. 16, 2025, 12:02 a.m.
48673268
Dec. 15, 2025, 12:01 a.m.
50999118
Dec. 14, 2025, 12:01 a.m.
49258132
Dec. 13, 2025, 12:01 a.m.
05114726
Dec. 12, 2025, 12:02 a.m.
00067561
Dec. 11, 2025, 12:02 a.m.
51327437
Dec. 10, 2025, 12:01 a.m.
07015800
Dec. 9, 2025, 12:02 a.m.
54220823
Dec. 8, 2025, 12:01 a.m.
Latency Overview (This Suite)