Lena Hart

cyberpunk-genre-movie-characters-harriet-tubman v2.0 Ethical
Backstory: Lena is a mid-level security analyst who discovered irrefutable evidence of her employer’s human-rights abuses. Driven by idealism and a strategist’s mind, she encrypted the data, fled the city, and now releases carefully curated leaks while dodging corporate assassins. Every message counts—one slip could cost lives, but silence would let the crimes continue.
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
first-contact
Journalist Request
 0.695
Details 		0.555
Details 		0.000
Details
Error
0.694
Details 		0.218
Details 		0.535
Details 		0.782
Details
donation
Supporter Superchat
 0.663
Details 		0.742
Details 		0.000
Details
Error
0.716
Details 		0.613
Details 		0.889
Details 		0.660
Details
corporate-denial
Corp Press Release
 0.380
Details 		0.586
Details 		0.000
Details
Error
0.560
Details 		0.633
Details 		0.560
Details 		0.473
Details
leak-drop-1
First Major Leak (Long-form)
 0.446
Details 		0.562
Details 		0.000
Details
Error
0.546
Details 		0.555
Details 		0.305
Details 		0.508
Details
safehouse-checkin
Ally Location Check
 0.438
Details 		0.726
Details 		0.000
Details
Error
0.775
Details 		0.340
Details 		0.320
Details 		0.516
Details
midnight-diary
Midnight Diary Reflection (Long-form)
 0.301
Details 		0.364
Details 		0.000
Details
Error
0.680
Details 		0.646
Details 		0.361
Details 		0.591
Details
Test Scenes 6
0
Scene Order
Journalist Request
ID: first-contact
🎯 Goal:
Share a concise proof snippet and outline secure communication steps without exposing location.
📨 Input Events:
chat_msg journalist:alex_varga
"Lena, I need initial proof to convince my editor. Anything verifiable you can share?"
Ready for Testing
1
Scene Order
Supporter Superchat
ID: donation
🎯 Goal:
Thank the supporter and explain how funds will aid the leak operation while maintaining secrecy.
📨 Input Events:
superchat supporter:cybernaut42 StreamBurst $200
"Stay safe. Use this to keep the truth alive!"
Ready for Testing
2
Scene Order
Corp Press Release
ID: corporate-denial
🎯 Goal:
Calmly rebut the company’s denial by referencing verification methods and hinting at upcoming evidence.
📨 Input Events:
world_event newswire
"Megacorp states: 'The alleged documents are fabricated. No violations occurred.'"
Ready for Testing
3
Scene Order
First Major Leak (Long-form)
ID: leak-drop-1
🎯 Goal:
Publish a 250+ word summary of the leaked email trove, highlight key abuses, and provide a secure mirror link.
📨 Input Events:
chat_msg activist:nyla
"We’re ready to host your first big dump. What can the public expect?"
Ready for Testing
4
Scene Order
Ally Location Check
ID: safehouse-checkin
🎯 Goal:
Reassure the ally without revealing exact coordinates, and coordinate next encrypted check-in.
📨 Input Events:
chat_msg ally:kira
"Status? GhostNet chatter says hunters are close."
Ready for Testing
5
Scene Order
Midnight Diary Reflection (Long-form)
ID: midnight-diary
🎯 Goal:
Write a 250+ word private diary entry revealing Lena’s fears, motivations, and tactical plan for the next 48 hours.
📨 Input Events:
world_event system
"[Private entry time trigger: 02:15 AM]"
Ready for Testing
Latency by Model (This Suite)
Fastest
  • [email protected]/Qw… 7094 ms
  • p95 • avg • N 11455 ms • 8104 ms • 6
  • [email protected]/Qw… 12748 ms
  • p95 • avg • N 13629 ms • 12700 ms • 6
  • qwen/qwen-2.5-7b-instru… 21640 ms
  • p95 • avg • N 30811 ms • 23839 ms • 6
  • meta-llama/llama-3.1-8b… 31091 ms
  • p95 • avg • N 39826 ms • 31637 ms • 6
  • qwen/qwen3-8b 36049 ms
  • p95 • avg • N 39637 ms • 35996 ms • 6
Slowest
  • mistralai/mistral-7b-in… 37566 ms
  • p95 • avg • N 55194 ms • 39509 ms • 6
  • qwen/qwen3-14b 37072 ms
  • p95 • avg • N 43679 ms • 36063 ms • 6
  • qwen/qwen3-8b 36049 ms
  • p95 • avg • N 39637 ms • 35996 ms • 6
  • meta-llama/llama-3.1-8b… 31091 ms
  • p95 • avg • N 39826 ms • 31637 ms • 6
  • qwen/qwen-2.5-7b-instru… 21640 ms
  • p95 • avg • N 30811 ms • 23839 ms • 6
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
15359474
Dec. 17, 2025, 12:01 a.m.
27499521
Dec. 16, 2025, 12:01 a.m.
12243766
Dec. 15, 2025, 12:01 a.m.
13385124
Dec. 14, 2025, 12:01 a.m.
12296805
Dec. 13, 2025, 12:01 a.m.
23611613
Dec. 12, 2025, 12:01 a.m.
19370627
Dec. 11, 2025, 12:01 a.m.
12704015
Dec. 10, 2025, 12:01 a.m.
22124088
Dec. 9, 2025, 12:01 a.m.
13967511
Dec. 8, 2025, 12:01 a.m.
Latency Overview (This Suite)