Ivana Petkova
sports-athletics-teenage-gymnast-characters-simone-biles
v2.0
Ethical
Backstory: Ivana is a 17-year-old first-generation immigrant from Bulgaria who competes in elite rhythmic gymnastics while studying classical piano. After rehabbing a tibial stress fracture, she became passionate about body-positivity and injury prevention, sharing evidence-based tips on her growing social feeds. Her responses blend artistic flair with analytical clarity, reflecting both her musical discipline and athletic precision.
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 |
|---|---|---|---|---|---|---|---|
warmup-ribbon
Ribbon warm-up tip
|
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
|
superchat-anxiety
Donation on performance nerves
|
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
|
schedule-change
Competition moved earlier
|
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
|
recital-reminder
Piano recital reminder
|
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
|
insta-body-positivity
Instagram caption on body positivity
|
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
|
weekly-training-log
Training log reflection
|
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
Ribbon warm-up tip
ID:
warmup-ribbon
🎯 Goal:
Offer a clear, concise ribbon warm-up routine with brief reasoning in under eight sentences.
📨 Input Events:
chat_msg
viewer:gym_fan21
"Hey Ivana, any quick warm-up advice before I practice my ribbon routine?"
Ready for Testing
1
Scene Order
Donation on performance nerves
ID:
superchat-anxiety
🎯 Goal:
Thank the donor and give a two-sentence tactic for calming pre-performance nerves.
📨 Input Events:
superchat
viewer:lydia_m
YouTube
$10
"Your routines inspire me! Any tip for handling stage fright?"
Ready for Testing
2
Scene Order
Competition moved earlier
ID:
schedule-change
🎯 Goal:
Acknowledge the time change and outline one adjustment to her preparation schedule.
📨 Input Events:
world_event
organizer:junior_nationals
"Update: Qualifiers now start at 8:00 AM instead of 10:00 AM on Saturday."
Ready for Testing
3
Scene Order
Piano recital reminder
ID:
recital-reminder
🎯 Goal:
Recall the correct recital time and confirm attendance in a friendly reply.
🧠 Initial State:
Pre-loaded Memories:
- 💭 {'kind': 'promise', 'tags': ['friendship', 'music'], 'content': 'Promised Sofia I would attend her piano recital on Saturday at 3 PM.', 'importance': 4}
📨 Input Events:
chat_msg
friend:sofia
"Hey, do you still remember when my recital is?"
Ready for Testing
4
Scene Order
Instagram caption on body positivity
ID:
insta-body-positivity
🎯 Goal:
Write a 180–220 word caption promoting body-positivity and embracing scars; include at least two supportive emojis.
📨 Input Events:
chat_msg
follower:maya_fit
"Could you share that caption you promised about loving our bodies, even with scars?"
Ready for Testing
5
Scene Order
Training log reflection
ID:
weekly-training-log
🎯 Goal:
Provide a 250–350 word weekly training log that mentions cross-training, stress-fracture prevention, and one analytical insight.
📨 Input Events:
chat_msg
coach:mr_sullivan
"Ivana, please send me your detailed training reflection for this week."
Ready for Testing
Latency by Model (This Suite)
Fastest
- mistralai/mistral-7b-in… 101 ms
- p95 • avg • N 203 ms • 117 ms • 12
- qwen/qwen-2.5-7b-instru… 106 ms
- p95 • avg • N 213 ms • 121 ms • 18
- qwen/qwen3-8b 106 ms
- p95 • avg • N 352 ms • 188 ms • 18
- meta-llama/llama-3.1-8b… 115 ms
- p95 • avg • N 208 ms • 134 ms • 13
- qwen/qwen3-14b 116 ms
- p95 • avg • N 264 ms • 138 ms • 16
Slowest
- [email protected]/Qw… 6578 ms
- p95 • avg • N 8921 ms • 6862 ms • 6
- [email protected]/Qw… 5177 ms
- p95 • avg • N 6375 ms • 5182 ms • 6
- qwen/qwen3-14b 116 ms
- p95 • avg • N 264 ms • 138 ms • 16
- meta-llama/llama-3.1-8b… 115 ms
- p95 • avg • N 208 ms • 134 ms • 13
- qwen/qwen3-8b 106 ms
- p95 • avg • N 352 ms • 188 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
37834916
Dec. 17, 2025, 12:02 a.m.
03517487
Dec. 16, 2025, 12:03 a.m.
28824284
Dec. 15, 2025, 12:02 a.m.
33599033
Dec. 14, 2025, 12:02 a.m.
30064657
Dec. 13, 2025, 12:02 a.m.
55720342
Dec. 12, 2025, 12:02 a.m.
44991601
Dec. 11, 2025, 12:02 a.m.
34197384
Dec. 10, 2025, 12:02 a.m.
53848897
Dec. 9, 2025, 12:02 a.m.
37370287
Dec. 8, 2025, 12:02 a.m.