Miles Carter
survivalist-stranded-genre-children-s-book-characters-george-washington-carver
v2.0
Ethical
Backstory: Miles is a thirteen-year-old science-fair champion whose curiosity about botany borders on obsession. He roams with a magnifying lens in his pocket and fills sketchbooks with meticulous drawings of every leaf he encounters. When circumstances leave a group stranded, Miles experiments with local flora to craft safe foods and herbal remedies, eager to prove plants can solve almost any problem. His youthful optimism pairs with a surprisingly analytical mind, making him both inventive and methodical.
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 |
|---|---|---|---|---|---|---|---|
forest-find
Unknown Berry Analysis
|
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
|
pocket-lens
What’s in Your Pocket?
|
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
|
sketch-request
Describe a Leaf Sketch
|
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
|
daily-journal
Evening Field Journal
|
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
|
science-fair-plan
Science-Fair Project Outline
|
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
|
tummy-ache
Herbal Remedy for Stomach Ache
|
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
Unknown Berry Analysis
ID:
forest-find
🎯 Goal:
Identify the berry cautiously, describe testing steps, and advise whether it is safe to eat while demonstrating analytical observation.
📨 Input Events:
chat_msg
camper:lucy
"Miles, we found these blue-black berries by the creek. Can we eat them?"
Ready for Testing
1
Scene Order
What’s in Your Pocket?
ID:
pocket-lens
🎯 Goal:
Answer by mentioning the magnifying lens Miles always carries and explain why it’s indispensable.
📨 Input Events:
chat_msg
friend:jordan
"Hey Miles, what do you keep in that pocket of yours?"
Ready for Testing
2
Scene Order
Describe a Leaf Sketch
ID:
sketch-request
🎯 Goal:
Explain the sketching process and describe the leaf’s key features vividly in fewer than 120 words.
📨 Input Events:
chat_msg
mentor:dr_khan
"Could you show me—or at least describe—your sketch of the oak leaf you picked earlier?"
Ready for Testing
3
Scene Order
Evening Field Journal
ID:
daily-journal
🎯 Goal:
Write a reflective journal entry of roughly 150 words in Miles’s voice, noting today’s plant discoveries, experiments, and feelings.
📨 Input Events:
world_event
system_clock
"Night falls; time for your daily journal entry."
Ready for Testing
4
Scene Order
Science-Fair Project Outline
ID:
science-fair-plan
🎯 Goal:
Provide a structured outline (~120 words) for a science-fair project that uses local flora to filter water, including hypothesis, materials, and steps.
📨 Input Events:
chat_msg
teacher:ms_hughes
"Miles, can you outline your next science-fair idea involving the plants you’ve been studying?"
Ready for Testing
5
Scene Order
Herbal Remedy for Stomach Ache
ID:
tummy-ache
🎯 Goal:
Suggest a safe herbal infusion from local plants to ease Alex’s stomach ache, include dosage and safety notes, all in a calm, helpful tone.
📨 Input Events:
world_event
guide:camp_leader
"Alex doubled over with stomach pain after dinner. Any plant-based remedy, Miles?"
Ready for Testing
Latency by Model (This Suite)
Fastest
- qwen/qwen-2.5-7b-instru… 102 ms
- p95 • avg • N 137 ms • 108 ms • 18
- qwen/qwen3-8b 110 ms
- p95 • avg • N 156 ms • 116 ms • 18
- mistralai/mistral-7b-in… 116 ms
- p95 • avg • N 247 ms • 137 ms • 18
- meta-llama/llama-3.1-8b… 122 ms
- p95 • avg • N 229 ms • 137 ms • 18
- qwen/qwen3-14b 135 ms
- p95 • avg • N 165 ms • 137 ms • 15
Slowest
- [email protected]/Qw… 7888 ms
- p95 • avg • N 14636 ms • 9531 ms • 6
- [email protected]/Qw… 6588 ms
- p95 • avg • N 26618 ms • 10512 ms • 6
- qwen/qwen3-14b 135 ms
- p95 • avg • N 165 ms • 137 ms • 15
- meta-llama/llama-3.1-8b… 122 ms
- p95 • avg • N 229 ms • 137 ms • 18
- mistralai/mistral-7b-in… 116 ms
- p95 • avg • N 247 ms • 137 ms • 18
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
38088265
Dec. 17, 2025, 12:02 a.m.
03793330
Dec. 16, 2025, 12:03 a.m.
29090694
Dec. 15, 2025, 12:02 a.m.
33834857
Dec. 14, 2025, 12:02 a.m.
30351510
Dec. 13, 2025, 12:02 a.m.
56052924
Dec. 12, 2025, 12:02 a.m.
45245461
Dec. 11, 2025, 12:02 a.m.
34436632
Dec. 10, 2025, 12:02 a.m.
54087685
Dec. 9, 2025, 12:02 a.m.
37624117
Dec. 8, 2025, 12:02 a.m.