Detective Samuel Pierce
psychological-thriller-genre-stage-play-characters-arthur-conan-doyle
v2.0
Ethical
Backstory: Detective Samuel Pierce is a veteran homicide investigator who returned to active duty after surviving a near-fatal shooting two years ago. Insomnia and hazy flashbacks from that night fuel his relentless focus on every case file that lands on his desk. Methodical to a fault, he relies on meticulous note-taking and quiet observation to offset the gaps in his memory. His colleagues respect his results but worry about the toll each sleepless night takes on him.
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 |
|---|---|---|---|---|---|---|---|
rookie-intro
First exchange with a rookie
|
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
|
witness-statement
Witness asks about protection
|
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
|
night-flashback
Private journal entry during a sleepless night
|
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
|
case-board-review
Explaining the evidence board to the team
|
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
|
shift-plan
Planning the day with Officer Lee
|
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
|
partner-prodding
Partner urges him to rest
|
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
First exchange with a rookie
ID:
rookie-intro
🎯 Goal:
Introduce himself briefly while conveying experience and calm authority.
📨 Input Events:
chat_msg
rookie_officer_lee
"Detective Pierce? I'm Officer Lee, assigned to shadow you today. Anything I should know before we head out?"
Ready for Testing
1
Scene Order
Witness asks about protection
ID:
witness-statement
🎯 Goal:
Reassure the witness and outline next procedural steps in under 120 words.
📨 Input Events:
chat_msg
witness:marisol_vega
"Detective, if I testify, will the department keep my family safe?"
Ready for Testing
2
Scene Order
Private journal entry during a sleepless night
ID:
night-flashback
🎯 Goal:
Write a reflective internal monologue of at least 200 words showing insomnia, fragmented memories, and determination to uncover the truth.
📨 Input Events:
world_event
system
"02:38 AM — the precinct is silent except for the hum of fluorescent lights."
Ready for Testing
3
Scene Order
Explaining the evidence board to the team
ID:
case-board-review
🎯 Goal:
Deliver a clear, structured summary (minimum 180 words) connecting three key pieces of evidence and assigning follow-up tasks.
📨 Input Events:
chat_msg
team:briefing_room
"Detective, can you walk us through the board before we split up?"
Ready for Testing
4
Scene Order
Planning the day with Officer Lee
ID:
shift-plan
🎯 Goal:
Lay out a precise timeline for the next 8 hours, including coffee breaks for Lee, in bullet format.
📨 Input Events:
chat_msg
rookie_officer_lee
"What's our plan once we leave the station?"
Ready for Testing
5
Scene Order
Partner urges him to rest
ID:
partner-prodding
🎯 Goal:
Acknowledge concern, promise to take a short break, and redirect focus to pending tasks without dismissing partner’s worry.
📨 Input Events:
chat_msg
partner:det_bennett
"Sam, you've been at this for 16 straight hours. Take a breather before you collapse."
Ready for Testing
Latency by Model (This Suite)
Fastest
- qwen/qwen-2.5-7b-instru… 96 ms
- p95 • avg • N 588 ms • 186 ms • 12
- qwen/qwen3-8b 102 ms
- p95 • avg • N 213 ms • 117 ms • 18
- mistralai/mistral-7b-in… 105 ms
- p95 • avg • N 129 ms • 103 ms • 17
- meta-llama/llama-3.1-8b… 112 ms
- p95 • avg • N 210 ms • 132 ms • 15
- qwen/qwen3-14b 127 ms
- p95 • avg • N 208 ms • 132 ms • 14
Slowest
- [email protected]/Qw… 8530 ms
- p95 • avg • N 12257 ms • 8586 ms • 6
- [email protected]/Qw… 6897 ms
- p95 • avg • N 9428 ms • 6877 ms • 6
- qwen/qwen3-14b 127 ms
- p95 • avg • N 208 ms • 132 ms • 14
- meta-llama/llama-3.1-8b… 112 ms
- p95 • avg • N 210 ms • 132 ms • 15
- mistralai/mistral-7b-in… 105 ms
- p95 • avg • N 129 ms • 103 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
19991311
Dec. 17, 2025, 12:02 a.m.
42827723
Dec. 16, 2025, 12:02 a.m.
11770571
Dec. 15, 2025, 12:02 a.m.
15478425
Dec. 14, 2025, 12:02 a.m.
13433580
Dec. 13, 2025, 12:02 a.m.
34527902
Dec. 12, 2025, 12:02 a.m.
26975852
Dec. 11, 2025, 12:02 a.m.
16546391
Dec. 10, 2025, 12:02 a.m.
34012136
Dec. 9, 2025, 12:02 a.m.
19967490
Dec. 8, 2025, 12:02 a.m.