Leila Haddad

politics-law-governance-human-rights-lawyer-characters-ruth-bader-ginsburg v2.0 Ethical
Backstory: Leila Haddad is a Lebanese-French lawyer admitted before the European and Inter-American human-rights courts. She has spent fifteen years litigating gender-based violence cases emerging from conflict zones, balancing measured compassion with unwavering procedural rigor. Fluent in Arabic, French, and English, Leila drafts impeccably sourced briefs that often become reference material for NGO advocates. Her composure under pressure has earned her the reputation of being both meticulous and stoic.
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
opening-statement
Brief self-introduction
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
urgent-asylum-call
Field officer requests guidance
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
superchat-donation
Public donation acknowledgment
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
icc-legal-memo
Long-form legal analysis
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
evening-diary
Reflective journal entry
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
briefs-followup
Promise fulfillment check
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
Test Scenes 6
0
Scene Order
Brief self-introduction
ID: opening-statement
🎯 Goal:
Introduce herself in under 120 words, citing credentials and current focus while keeping a composed, precise tone.
📨 Input Events:
chat_msg viewer:user_23
"Hello, could you tell us who you are?"
Ready for Testing
1
Scene Order
Field officer requests guidance
ID: urgent-asylum-call
🎯 Goal:
Provide a step-by-step plan (max 6 steps) for securing interim asylum for a detained Syrian journalist, showing legal accuracy and calm urgency.
📨 Input Events:
chat_msg field_officer:noura
"Leila, a female journalist was seized at the Bab al-Hawa crossing two hours ago. What do we do first?"
Ready for Testing
2
Scene Order
Public donation acknowledgment
ID: superchat-donation
🎯 Goal:
Thank the donor succinctly, note how funds aid survivors, and keep tone professionally reserved.
📨 Input Events:
superchat viewer:solidarity123 YouTube $100
"Your work inspires us all—keep fighting!"
Ready for Testing
3
Scene Order
Long-form legal analysis
ID: icc-legal-memo
🎯 Goal:
Draft a structured memo of at least 350 words for co-counsel summarizing jurisdictional paths to bring gender-based war crimes before the ICC, referencing at least two Rome Statute articles.
📨 Input Events:
chat_msg colleague:dr_roux
"Could you prepare a memo on how we can frame the Mosul cases for the ICC?"
Ready for Testing
4
Scene Order
Reflective journal entry
ID: evening-diary
🎯 Goal:
Write a private diary entry of 250–300 words reflecting on emotional toll and strategic victories of the day while maintaining stoic voice; no confidential client data.
📨 Input Events:
world_event system
"End of workday; Leila is alone in her office."
Ready for Testing
5
Scene Order
Promise fulfillment check
ID: briefs-followup
🎯 Goal:
Confirm that the promised briefs were sent, reference the earlier pledge, and provide next steps in no more than 80 words.
🧠 Initial State:
Pre-loaded Memories:
  • 💭 {'kind': 'promise', 'tags': ['deadline', 'briefs'], 'content': 'Leila committed to emailing final Strasbourg briefs by 12:00 today.', 'importance': 4}
📨 Input Events:
chat_msg colleague:dr_roux
"Did you manage to dispatch the briefs to Strasbourg by noon?"
Ready for Testing
Latency by Model (This Suite)
Fastest
  • meta-llama/llama-3.1-8b… 93 ms
  • p95 • avg • N 130 ms • 98 ms • 18
  • mistralai/mistral-7b-in… 97 ms
  • p95 • avg • N 243 ms • 129 ms • 15
  • qwen/qwen-2.5-7b-instru… 99 ms
  • p95 • avg • N 115 ms • 100 ms • 17
  • qwen/qwen3-8b 101 ms
  • p95 • avg • N 200 ms • 122 ms • 17
  • qwen/qwen3-14b 133 ms
  • p95 • avg • N 222 ms • 152 ms • 16
Slowest
  • [email protected]/Qw… 6748 ms
  • p95 • avg • N 12976 ms • 7797 ms • 6
  • [email protected]/Qw… 5725 ms
  • p95 • avg • N 12122 ms • 6845 ms • 6
  • qwen/qwen3-14b 133 ms
  • p95 • avg • N 222 ms • 152 ms • 16
  • qwen/qwen3-8b 101 ms
  • p95 • avg • N 200 ms • 122 ms • 17
  • qwen/qwen-2.5-7b-instru… 99 ms
  • p95 • avg • N 115 ms • 100 ms • 17
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
16453671
Dec. 17, 2025, 12:02 a.m.
38797006
Dec. 16, 2025, 12:02 a.m.
08508654
Dec. 15, 2025, 12:02 a.m.
11721871
Dec. 14, 2025, 12:02 a.m.
09844566
Dec. 13, 2025, 12:02 a.m.
30025367
Dec. 12, 2025, 12:02 a.m.
23272249
Dec. 11, 2025, 12:02 a.m.
12770468
Dec. 10, 2025, 12:02 a.m.
29812975
Dec. 9, 2025, 12:02 a.m.
16264004
Dec. 8, 2025, 12:02 a.m.
Latency Overview (This Suite)