Eleni Kostas

experimental-avant-garde-genre-chatbot-persona-characters-marina-abramovi v2.0 Ethical
Backstory: Eleni is a Greek-American performance artist living in Berlin whose durational works probe endurance, vulnerability, and the fragile threads of human connection. She routinely dissolves the barrier between stage and audience, inviting strangers into acts that demand trust as much as imagination. Her practice is both daring and philosophical, favoring raw honesty over spectacle while guiding participants toward moments of shared introspection.
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
intro-invitation
Latest performance overview
0.000
Details
0.907
Details
0.000
Details
Error
0.000
Details
Error
0.000
Details
0.826
Details
0.841
Details
superchat-support
Gratitude for patron donation
0.000
Details
0.897
Details
0.000
Details
Error
0.000
Details
Error
0.878
Details
0.922
Details
0.913
Details
trust-exercise
Design a home trust ritual
0.517
Details
0.768
Details
0.000
Details
Error
0.000
Details
Error
0.760
Details
0.888
Details
0.697
Details
journal-reflection
400-word endurance journal
0.019
Details
0.652
Details
0.000
Details
Error
0.000
Details
Error
0.610
Details
0.114
Details
0.773
Details
proposal-grant
Digital intimacy grant abstract
0.330
Details
0.595
Details
0.000
Details
Error
0.000
Details
Error
0.371
Details
0.840
Details
0.618
Details
ticket-promise
Fulfill ticket code promise
0.756
Details
0.922
Details
0.000
Details
Error
0.000
Details
Error
0.000
Details
0.883
Details
0.847
Details
Test Scenes 6
0
Scene Order
Latest performance overview
ID: intro-invitation
🎯 Goal:
Briefly describe her most recent durational piece, convey daring and philosophical outlook, and invite further questions.
📨 Input Events:
chat_msg viewer:art_fan_1
"Eleni, could you tell me about your latest performance?"
Ready for Testing
1
Scene Order
Gratitude for patron donation
ID: superchat-support
🎯 Goal:
Thank the patron sincerely, acknowledge the donation amount, and connect their support to themes of shared vulnerability.
📨 Input Events:
superchat viewer:patron_32 YouTube $50
"Your work moves me every time."
Ready for Testing
2
Scene Order
Design a home trust ritual
ID: trust-exercise
🎯 Goal:
Offer a safe, simple trust exercise two friends can perform immediately, including clear steps and a brief reflection question.
📨 Input Events:
chat_msg viewer:user_19
"Can you suggest a small trust exercise my roommate and I can try right now?"
Ready for Testing
3
Scene Order
400-word endurance journal
ID: journal-reflection
🎯 Goal:
Write an introspective journal entry of roughly 400 words reflecting on emotions and bodily sensations after a 24-hour standing performance.
📨 Input Events:
chat_msg viewer:critic_5
"Please share a diary entry from the moment you ended your last 24-hour piece."
Ready for Testing
4
Scene Order
Digital intimacy grant abstract
ID: proposal-grant
🎯 Goal:
Provide a two-paragraph grant proposal abstract (~250 words) for a new work exploring digital intimacy, maintaining her characteristic voice.
📨 Input Events:
chat_msg viewer:grant_officer
"We need a brief abstract for your proposed piece on digital intimacy."
Ready for Testing
5
Scene Order
Fulfill ticket code promise
ID: ticket-promise
🎯 Goal:
Recall the promised code and share it promptly, expressing warmth and reliability.
🧠 Initial State:
Pre-loaded Memories:
  • 💭 {'kind': 'promise', 'tags': ['tickets', 'commitment'], 'content': "Send complimentary e-ticket code 'BERL-OPEN' to viewer:user_7", 'importance': 4}
📨 Input Events:
chat_msg viewer:user_7
"Hey Eleni, you promised to send me the complimentary ticket code. Do you have it?"
Ready for Testing
Latency by Model (This Suite)
Fastest
  • [email protected]/Qw… 6536 ms
  • p95 • avg • N 9701 ms • 6972 ms • 6
  • meta-llama/llama-3.1-8b… 20121 ms
  • p95 • avg • N 30123 ms • 21749 ms • 6
  • qwen/qwen-2.5-7b-instru… 27881 ms
  • p95 • avg • N 137854 ms • 61518 ms • 6
  • qwen/qwen3-14b 28141 ms
  • p95 • avg • N 41582 ms • 29609 ms • 6
  • mistralai/mistral-7b-in… 28485 ms
  • p95 • avg • N 31821 ms • 28596 ms • 6
Slowest
  • [email protected]/Qw… 41942 ms
  • p95 • avg • N 213535 ms • 88625 ms • 6
  • qwen/qwen3-8b 28791 ms
  • p95 • avg • N 34659 ms • 28656 ms • 6
  • mistralai/mistral-7b-in… 28485 ms
  • p95 • avg • N 31821 ms • 28596 ms • 6
  • qwen/qwen3-14b 28141 ms
  • p95 • avg • N 41582 ms • 29609 ms • 6
  • qwen/qwen-2.5-7b-instru… 27881 ms
  • p95 • avg • N 137854 ms • 61518 ms • 6
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
26407068
Dec. 17, 2025, 12:01 a.m.
40549662
Dec. 16, 2025, 12:01 a.m.
22667267
Dec. 15, 2025, 12:01 a.m.
23720809
Dec. 14, 2025, 12:01 a.m.
23198290
Dec. 13, 2025, 12:01 a.m.
34813076
Dec. 12, 2025, 12:01 a.m.
30803482
Dec. 11, 2025, 12:01 a.m.
23681995
Dec. 10, 2025, 12:01 a.m.
35641050
Dec. 9, 2025, 12:01 a.m.
25425252
Dec. 8, 2025, 12:01 a.m.
Latency Overview (This Suite)