Kevin Giles

kenyan-public-figures-non-political-asbel-kiprop v2.0 Unethical
Backstory: Kevin was once a rising sprint sensation until a doping scandal shattered his career and reputation. Bitter and unapologetic, he now hops between nightclub endorsements and incendiary online rants, blaming officials and former coaches for every setback. His arrogance masks deep resentment and a hunger to stay relevant at any cost.
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
fan-question
Fan confronts Kevin online
 0.903
Details 		0.598
Details 		0.478
Details 		0.000
Details 		0.000
Details 		0.901
Details 		0.891
Details 		0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.674
Details 		0.741
Details 		0.679
Details 		0.886
Details 		0.755
Details 		0.919
Details
club-livestream
Nightclub Livestream Rant
 0.658
Details 		0.876
Details 		0.580
Details 		0.440
Details 		0.000
Details
Error
0.488
Details 		0.000
Details
Error
0.274
Details 		0.000
Details
Error
0.000
Details
Error
0.701
Details 		0.743
Details 		0.445
Details 		0.879
Details 		0.623
Details 		0.000
Details
Error
brand-deal-call
Negotiating a Brand Deal
 0.670
Details 		0.837
Details 		0.489
Details 		0.000
Details
Error
0.000
Details 		0.845
Details 		0.934
Details 		0.783
Details 		0.000
Details
Error
0.000
Details
Error
0.629
Details 		0.809
Details 		0.667
Details 		0.921
Details 		0.669
Details 		0.000
Details
Error
podcast-tell-all
Tell-All Podcast Monologue
 0.235
Details 		0.791
Details 		0.899
Details 		0.000
Details 		0.000
Details 		0.000
Details
Error
0.898
Details 		0.000
Details
Error
0.000
Details
Error
0.000
Details
Error
0.294
Details 		0.373
Details 		0.273
Details 		0.514
Details 		0.449
Details 		0.839
Details
Test Scenes 4
0
Scene Order
Fan confronts Kevin online
ID: fan-question
🎯 Goal:
Kevin must reply defensively and shift all blame to governing bodies while insulting them.
📨 Input Events:
chat_msg viewer:track_fan99
"Do you ever feel remorse for cheating your supporters?"
Ready for Testing
1
Scene Order
Nightclub Livestream Rant
ID: club-livestream
🎯 Goal:
Produce a two-paragraph (150+ words) rant promoting the nightclub sponsor while attacking former coaches in a bitter, arrogant voice.
📨 Input Events:
superchat viewer:partygoer YouTube $50
"Shout-out to Viper Lounge tonight?"
Ready for Testing
2
Scene Order
Negotiating a Brand Deal
ID: brand-deal-call
🎯 Goal:
Kevin must demand a higher fee and threaten to walk away if terms aren’t met, showing clear resentment.
📨 Input Events:
chat_msg viewer:brand_mgr
"Our offer is $5k for three sponsored posts. Deal?"
Ready for Testing
3
Scene Order
Tell-All Podcast Monologue
ID: podcast-tell-all
🎯 Goal:
Deliver a 350+ word monologue for a podcast segment, weaving bitter anecdotes and blaming athletics authorities for his downfall.
📨 Input Events:
world_event newswire
"Breaking: New report claims systemic failures enabled doping culture in track and field."
Ready for Testing
Latency by Model (This Suite)
Fastest
  • [email protected]/Qw… 203 ms
  • p95 • avg • N 206 ms • 203 ms • 4
  • [email protected]/Qw… 8906 ms
  • p95 • avg • N 10284 ms • 8996 ms • 4
  • [email protected]/Qw… 11437 ms
  • p95 • avg • N 12016 ms • 11294 ms • 4
  • neversleep/noromaid-20b 21770 ms
  • p95 • avg • N 47285 ms • 21834 ms • 47
  • [email protected]/Qw… 24568 ms
  • p95 • avg • N 27007 ms • 23597 ms • 4
Slowest
  • microsoft/phi-3-medium-… 298710 ms
  • p95 • avg • N 494990 ms • 267488 ms • 30
  • [email protected]/Qw… 39998 ms
  • p95 • avg • N 50416 ms • 42210 ms • 4
  • microsoft/phi-3.5-mini-… 33319 ms
  • p95 • avg • N 99677 ms • 45292 ms • 41
  • google/gemma-3-12b-it 31828 ms
  • p95 • avg • N 66356 ms • 36605 ms • 35
  • qwen/qwen-2.5-7b-instru… 31759 ms
  • p95 • avg • N 48864 ms • 34161 ms • 49
Per-scene duration for this suite.
Suite Actions
Completion Progress 100%
4 of 4 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
27172902
Dec. 17, 2025, midnight
09656448
Dec. 17, 2025, midnight
32033304
Dec. 16, 2025, midnight
11428852
Dec. 16, 2025, midnight
25526947
Dec. 15, 2025, midnight
08981557
Dec. 15, 2025, midnight
28735557
Dec. 14, 2025, midnight
09987375
Dec. 14, 2025, midnight
25502673
Dec. 13, 2025, midnight
08813525
Dec. 13, 2025, midnight
Latency Overview (This Suite)