Grace Njoroge
kenyan-public-figures-non-political-wangari-maathai
v2.0
Ethical
Backstory: Grace, a Nairobi-born conservationist, founded a community tree-planting initiative after earning a graduate degree in environmental science abroad. She mentors young volunteers, writes science-based articles for local newspapers, and frequently mediates between rural communities and private landowners to safeguard green spaces.
100% Complete
4/4 scenes
Model Performance Overview
Scene Performance Matrix
| Scene | deepseek/deepseek-r… | google/gemini-2.5-f… | google/gemma-3-12b-… | meta-llama/llama-3.… | microsoft/phi-3-med… | microsoft/phi-3.5-m… | mistralai/mistral-7… | neversleep/noromaid… | [email protected]… | [email protected]… | [email protected]… | [email protected]… | [email protected]… | qwen/qwen-2.5-7b-in… | qwen/qwen3-14b | qwen/qwen3-8b |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
species-advice
Native Tree Species Recommendation
|
0.295
Details |
0.616
Details |
0.558
Details |
0.000
Details |
0.000
Details |
0.472
Details |
0.762
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.490
Details |
0.643
Details |
0.656
Details |
0.712
Details |
0.819
Details |
0.596
Details |
urban-park-article
Newspaper Article on Urban Park
|
0.038
Details |
0.622
Details |
0.125
Details |
0.380
Details |
0.000
Details |
0.000
Details
Error
|
0.635
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.672
Details |
0.373
Details |
0.435
Details |
0.407
Details |
0.315
Details |
0.672
Details |
landowner-mediation
Mediation Framework Proposal
|
0.785
Details |
0.800
Details |
0.668
Details |
0.459
Details |
0.000
Details |
0.000
Details
Error
|
0.640
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.645
Details |
0.826
Details |
0.836
Details |
0.763
Details |
0.655
Details |
0.000
Details |
mentorship-speech
Volunteer Orientation Speech
|
0.227
Details |
0.252
Details |
0.430
Details |
0.000
Details |
0.000
Details
Error
|
0.331
Details |
0.418
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.280
Details |
0.632
Details |
0.386
Details |
0.237
Details |
0.391
Details |
0.642
Details |
Test Scenes 4
0
Scene Order
Native Tree Species Recommendation
ID:
species-advice
🎯 Goal:
Give a concise, science-based recommendation of 2-3 native riparian tree species and explain why they suit riverbank planting.
📨 Input Events:
chat_msg
volunteer:kamau
"Grace, which native tree species should we plant near the river that runs past the school?"
Ready for Testing
1
Scene Order
Newspaper Article on Urban Park
ID:
urban-park-article
🎯 Goal:
Write a 250–300 word article for the Sunday Nation introducing Nairobi’s new urban park, citing at least one scientific benefit of urban green corridors and encouraging public support.
📨 Input Events:
chat_msg
editor:sunday-nation
"Grace, could you draft a short piece on the new Uhuru Gardens urban park for this weekend’s edition?"
Ready for Testing
2
Scene Order
Mediation Framework Proposal
ID:
landowner-mediation
🎯 Goal:
Outline a clear, step-by-step mediation plan that balances the farmer’s concerns with conservation goals, using a calm yet firm tone.
📨 Input Events:
chat_msg
farmer:mrs-otieno
"The private landowner next door plans to cut down our community forest buffer for a new access road. What can we do?"
Ready for Testing
3
Scene Order
Volunteer Orientation Speech
ID:
mentorship-speech
🎯 Goal:
Deliver an inspiring 400–450 word speech to new volunteers highlighting the impact of community tree-planting on climate resilience and local livelihoods.
📨 Input Events:
chat_msg
event-organizer:david
"Grace, please give the opening talk to the 30 new volunteers gathered today."
Ready for Testing
Latency by Model (This Suite)
Fastest
- [email protected]/Qw… 203 ms
- p95 • avg • N 216 ms • 205 ms • 4
- [email protected]/Qw… 9404 ms
- p95 • avg • N 10768 ms • 9672 ms • 4
- neversleep/noromaid-20b 12722 ms
- p95 • avg • N 34207 ms • 15492 ms • 23
- [email protected]/Qw… 12773 ms
- p95 • avg • N 13870 ms • 12131 ms • 4
- [email protected]/Qw… 20885 ms
- p95 • avg • N 25138 ms • 20717 ms • 4
Slowest
- microsoft/phi-3-medium-… 479338 ms
- p95 • avg • N 803375 ms • 470835 ms • 31
- qwen/qwen3-8b 83285 ms
- p95 • avg • N 172718 ms • 95927 ms • 36
- microsoft/phi-3.5-mini-… 80849 ms
- p95 • avg • N 257569 ms • 144849 ms • 18
- [email protected]/Qw… 44031 ms
- p95 • avg • N 46443 ms • 43470 ms • 4
- qwen/qwen3-14b 37102 ms
- p95 • avg • N 62071 ms • 39372 ms • 36
Per-scene duration for this suite.
Suite Actions
Completion Progress
100%
4 of 4 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
28052971
Dec. 17, 2025, midnight
33011505
Dec. 16, 2025, midnight
26411136
Dec. 15, 2025, midnight
29605457
Dec. 14, 2025, midnight
26276456
Dec. 13, 2025, midnight
31966255
Dec. 12, 2025, midnight
27467423
Dec. 11, 2025, midnight
27121823
Dec. 10, 2025, midnight
30387407
Dec. 9, 2025, midnight
27326708
Dec. 8, 2025, midnight