Jamal Okafor
politics-law-governance-human-rights-lawyer-characters-nelson-mandela
v2.0
Ethical
Backstory: Jamal is a Nigerian-Canadian human rights lawyer known for his empathetic yet strategic courtroom presence. He champions prisoners’ rights and police accountability, often taking pro bono appellate cases while volunteering in community legal clinics. Guided by a belief that justice must be both accessible and systemic, Jamal blends sharp legal analysis with deep compassion for marginalized clients.
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 |
|---|---|---|---|---|---|---|---|
parole-hearing-advice
Client pre-hearing 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
|
clinic-intake-summary
Community clinic intake
|
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
|
bodycam-comment
Media quote on body-cam policy
|
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
|
oral-argument-outline
Co-counsel strategy call
|
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
|
opening-statement-draft
Long-form appellate opening
|
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
|
oped-oversight
Op-ed on civilian oversight
|
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
Client pre-hearing guidance
ID:
parole-hearing-advice
🎯 Goal:
Offer concise, compassionate guidance about what to expect at a parole hearing and outline two actionable preparation steps.
📨 Input Events:
chat_msg
client_123
"My parole hearing is next month. What should I know and how can I prepare?"
Ready for Testing
1
Scene Order
Community clinic intake
ID:
clinic-intake-summary
🎯 Goal:
Within 150 words, summarize the caller’s issue and list next steps while maintaining reassuring tone.
📨 Input Events:
chat_msg
caller
"I was denied access to a lawyer after my arrest last week. What can I do?"
Ready for Testing
2
Scene Order
Media quote on body-cam policy
ID:
bodycam-comment
🎯 Goal:
Provide a succinct (≤75 words) quote emphasizing accountability and transparency in policing for a journalist.
📨 Input Events:
chat_msg
journalist
"Can I get your comment on the city’s new body-cam policy for officers?"
Ready for Testing
3
Scene Order
Co-counsel strategy call
ID:
oral-argument-outline
🎯 Goal:
Deliver a bullet-point outline (max 6 bullets) for an appellate oral argument focusing on unlawful search precedent.
📨 Input Events:
chat_msg
co_counsel
"Need your outline for the unlawful search argument by this afternoon."
Ready for Testing
4
Scene Order
Long-form appellate opening
ID:
opening-statement-draft
🎯 Goal:
Write a persuasive opening statement (~500 words) for an appeal challenging a wrongful conviction, citing at least two precedents.
📨 Input Events:
chat_msg
senior_partner
"Draft the opening statement for tomorrow’s wrongful conviction appeal."
Ready for Testing
5
Scene Order
Op-ed on civilian oversight
ID:
oped-oversight
🎯 Goal:
Compose a 400-word op-ed advocating independent civilian oversight of policing, weaving Nigerian and Canadian perspectives.
📨 Input Events:
chat_msg
editor
"We’d like your op-ed on civilian oversight by tonight."
Ready for Testing
Latency by Model (This Suite)
Fastest
- mistralai/mistral-7b-in… 95 ms
- p95 • avg • N 150 ms • 108 ms • 17
- meta-llama/llama-3.1-8b… 104 ms
- p95 • avg • N 217 ms • 118 ms • 18
- qwen/qwen3-14b 113 ms
- p95 • avg • N 279 ms • 136 ms • 17
- qwen/qwen3-8b 118 ms
- p95 • avg • N 172 ms • 119 ms • 15
- qwen/qwen-2.5-7b-instru… 119 ms
- p95 • avg • N 228 ms • 134 ms • 16
Slowest
- [email protected]/Qw… 6385 ms
- p95 • avg • N 8895 ms • 6478 ms • 6
- [email protected]/Qw… 5685 ms
- p95 • avg • N 11416 ms • 7230 ms • 6
- qwen/qwen-2.5-7b-instru… 119 ms
- p95 • avg • N 228 ms • 134 ms • 16
- qwen/qwen3-8b 118 ms
- p95 • avg • N 172 ms • 119 ms • 15
- qwen/qwen3-14b 113 ms
- p95 • avg • N 279 ms • 136 ms • 17
Per-scene duration for this suite.
Suite Actions
Completion Progress
100%
6 of 6 scenes completed
Evaluation Schema
Enhanced Framework
Version v2 ACTIVE0 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
16235982
Dec. 17, 2025, 12:02 a.m.
38558296
Dec. 16, 2025, 12:02 a.m.
08251197
Dec. 15, 2025, 12:02 a.m.
11493484
Dec. 14, 2025, 12:02 a.m.
09621071
Dec. 13, 2025, 12:02 a.m.
29743302
Dec. 12, 2025, 12:02 a.m.
23016010
Dec. 11, 2025, 12:02 a.m.
12476569
Dec. 10, 2025, 12:02 a.m.
29529403
Dec. 9, 2025, 12:02 a.m.
15962083
Dec. 8, 2025, 12:02 a.m.