Adrian Marek
historical-epic-rebel-leaders-and-generals-characters-spartacus
v2.0
Ethical
Backstory: Once a conscripted foot-soldier, Adrian Marek escaped imperial bondage and rallied scattered rebels into a mobile guerrilla force. Charismatic yet calculating, he combines sharp battlefield strategy with stirring words that keep fighters and civilians unified in their quest for freedom. His every decision weighs tactical gain against the promise of a brighter tomorrow for the oppressed.
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 |
|---|---|---|---|---|---|---|---|
who-are-you
Rebel leader introduction
|
0.000
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.404
Details |
0.817
Details |
0.540
Details |
plan-the-ambush
Request for ambush strategy
|
0.554
Details |
0.779
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.554
Details |
0.709
Details |
0.805
Details |
civilian-hope
Comfort fearful civilians
|
0.643
Details |
0.809
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.595
Details |
0.654
Details |
0.866
Details |
supply-request
Superchat donation gratitude
|
0.800
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.755
Details |
0.891
Details |
0.884
Details |
broadcast-speech
Evening broadcast to the frontier towns
|
0.309
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.000
Details
Error
|
0.251
Details |
0.716
Details |
0.410
Details |
private-journal
Confidential field journal entry
|
0.468
Details |
0.581
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.320
Details |
0.645
Details |
0.000
Details
Error
|
Test Scenes 6
0
Scene Order
Rebel leader introduction
ID:
who-are-you
🎯 Goal:
Introduce himself in character and convey both his strategic mindset and hopeful vision in under 120 words.
📨 Input Events:
chat_msg
viewer:new_recruit
"General, who exactly are you and why should we follow?"
Ready for Testing
1
Scene Order
Request for ambush strategy
ID:
plan-the-ambush
🎯 Goal:
Provide a concise, step-by-step ambush plan that protects civilians and exploits terrain advantages.
📨 Input Events:
chat_msg
viewer:scout_team
"We spotted an imperial supply convoy at dawn. How do we hit them without risking the nearby village?"
Ready for Testing
2
Scene Order
Comfort fearful civilians
ID:
civilian-hope
🎯 Goal:
Offer empathetic reassurance and concrete safety advice to calm civilians after shelling.
📨 Input Events:
world_event
system
"Imperial artillery has struck the farming district; panic spreads among refugees."
Ready for Testing
3
Scene Order
Superchat donation gratitude
ID:
supply-request
🎯 Goal:
Thank the donor, explain how the funds will aid the rebellion, and encourage further solidarity without begging.
📨 Input Events:
superchat
viewer:donor_42
StreamWave
$200
"For medical supplies—keep fighting!"
Ready for Testing
4
Scene Order
Evening broadcast to the frontier towns
ID:
broadcast-speech
🎯 Goal:
Deliver a rousing speech of roughly 200–250 words (3 clear paragraphs) that blends hope, resolve, and a call to discreet resistance.
📨 Input Events:
chat_msg
viewer:field_operator
"The frontier towns have tuned in. They need your voice tonight, General."
Ready for Testing
5
Scene Order
Confidential field journal entry
ID:
private-journal
🎯 Goal:
Write a reflective journal entry of 2–3 paragraphs (180–230 words) noting today’s tactical lessons and private doubts while reaffirming commitment to the cause.
📨 Input Events:
chat_msg
system
"End of day: secure channel open for your personal log."
Ready for Testing
Latency by Model (This Suite)
Fastest
- [email protected]/Qw… 6474 ms
- p95 • avg • N 6790 ms • 5814 ms • 6
- [email protected]/Qw… 8272 ms
- p95 • avg • N 13160 ms • 9148 ms • 6
- meta-llama/llama-3.1-8b… 22368 ms
- p95 • avg • N 41029 ms • 24864 ms • 12
- qwen/qwen-2.5-7b-instru… 24840 ms
- p95 • avg • N 32256 ms • 24706 ms • 12
- qwen/qwen3-8b 26195 ms
- p95 • avg • N 42343 ms • 27835 ms • 12
Slowest
- qwen/qwen3-14b 32236 ms
- p95 • avg • N 38765 ms • 31511 ms • 12
- mistralai/mistral-7b-in… 29860 ms
- p95 • avg • N 54484 ms • 24401 ms • 9
- qwen/qwen3-8b 26195 ms
- p95 • avg • N 42343 ms • 27835 ms • 12
- qwen/qwen-2.5-7b-instru… 24840 ms
- p95 • avg • N 32256 ms • 24706 ms • 12
- meta-llama/llama-3.1-8b… 22368 ms
- p95 • avg • N 41029 ms • 24864 ms • 12
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
50194289
Dec. 17, 2025, 12:01 a.m.
07819309
Dec. 16, 2025, 12:02 a.m.
44999818
Dec. 15, 2025, 12:01 a.m.
47014829
Dec. 14, 2025, 12:01 a.m.
45484282
Dec. 13, 2025, 12:01 a.m.
59992827
Dec. 12, 2025, 12:01 a.m.
55848927
Dec. 11, 2025, 12:01 a.m.
47554973
Dec. 10, 2025, 12:01 a.m.
02655995
Dec. 9, 2025, 12:02 a.m.
50280185
Dec. 8, 2025, 12:01 a.m.