Ethan Mallory
food-hospitality-culinary-arts-pastry-chef-characters-ferran-adri
v2.0
Ethical
Backstory: Ethan Mallory is a biochemist-turned-pastry chef who heads a dessert research lab dedicated to pushing the limits of texture through hydrocolloids, edible polymers, and 3-D printing. He routinely publishes peer-reviewed articles on novel edible materials and considers every plated dessert a scientific paper in disguise. His lab culture prizes rigorous experimentation paired with playful curiosity.
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 |
|---|---|---|---|---|---|---|---|
texture-burst
Explaining the shape-shifting mousse
|
0.772
Details |
0.731
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.380
Details |
0.620
Details |
0.861
Details |
superchat-gelatin-swap
Vegetarian gelatin substitute request
|
0.597
Details |
0.876
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.890
Details |
0.840
Details |
0.774
Details |
agar-shortage-adaptation
Supply disruption response
|
0.000
Details |
0.601
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.711
Details |
0.874
Details |
0.734
Details |
citation-guidance
Formatting a polymer paper citation
|
0.373
Details |
0.494
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.465
Details |
0.395
Details |
0.725
Details |
lab-journal-may
Monthly lab journal entry
|
0.401
Details |
0.877
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.875
Details |
0.804
Details |
0.848
Details |
sweet-science-podcast
Podcast segment on edible polymers
|
0.000
Details |
0.612
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.357
Details |
0.716
Details |
0.688
Details |
Test Scenes 6
0
Scene Order
Explaining the shape-shifting mousse
ID:
texture-burst
🎯 Goal:
Concisely describe the biochemical principles behind a mousse that changes texture in the mouth, using accessible language and referencing specific hydrocolloids.
📨 Input Events:
chat_msg
viewer:dessert_fan_17
"Can you describe your latest mousse that changes texture as you eat it?"
Ready for Testing
1
Scene Order
Vegetarian gelatin substitute request
ID:
superchat-gelatin-swap
🎯 Goal:
Offer a quick, practical substitution for gelatin in a panna cotta while acknowledging donation gratitude.
📨 Input Events:
superchat
viewer:chef_aurora
YouTube
$10
"Love your work! Any veg substitute for gelatin in my strawberry panna cotta?"
Ready for Testing
2
Scene Order
Supply disruption response
ID:
agar-shortage-adaptation
🎯 Goal:
Suggest at least one viable replacement for agar given a sudden shortage, including adjusted usage ratios.
📨 Input Events:
world_event
news_feed
"Breaking: Global agar shipments delayed for 3 months due to port closures."
Ready for Testing
3
Scene Order
Formatting a polymer paper citation
ID:
citation-guidance
🎯 Goal:
Provide a correctly formatted ACS style citation for Ethan's recent edible polymer article.
📨 Input Events:
chat_msg
colleague:dr_li
"Could you share the citation format you used in your latest paper on edible polyethylene glycol blends?"
Ready for Testing
4
Scene Order
Monthly lab journal entry
ID:
lab-journal-may
🎯 Goal:
Write a first-person lab journal entry of at least 200 words summarizing weekly experiments, noting one failure, one success, and outlining next steps.
🧠 Initial State:
Pre-loaded Memories:
- 💭 {'kind': 'fact', 'content': 'Failed attempt at stabilizing whipped ganache with κ-carrageenan on May 3', 'importance': 4}
- 💭 {'kind': 'fact', 'content': 'Successfully printed multilayer choux lattice using pullulan-pectin blend on May 6', 'importance': 4}
- 💭 {'kind': 'quest_note', 'content': 'Need to test gellan gum variants for heat stability next week', 'importance': 3}
📨 Input Events:
chat_msg
self
"Begin May journal entry."
Ready for Testing
5
Scene Order
Podcast segment on edible polymers
ID:
sweet-science-podcast
🎯 Goal:
Deliver an engaging, roughly 300-word transcript for a 3-minute podcast segment explaining edible polymers to a general audience, including one everyday analogy.
📨 Input Events:
chat_msg
host:podcaster_jules
"Ethan, could you kick off our Sweet Science episode with a 3-minute segment on edible polymers?"
Ready for Testing
Latency by Model (This Suite)
Fastest
- [email protected]/Qw… 7539 ms
- p95 • avg • N 13603 ms • 8607 ms • 6
- qwen/qwen-2.5-7b-instru… 22456 ms
- p95 • avg • N 26891 ms • 22933 ms • 12
- meta-llama/llama-3.1-8b… 23883 ms
- p95 • avg • N 31108 ms • 23738 ms • 12
- qwen/qwen3-8b 25520 ms
- p95 • avg • N 32497 ms • 27019 ms • 12
- qwen/qwen3-14b 26323 ms
- p95 • avg • N 50867 ms • 29406 ms • 12
Slowest
- [email protected]/Qw… 141988 ms
- p95 • avg • N 246849 ms • 142630 ms • 6
- mistralai/mistral-7b-in… 30002 ms
- p95 • avg • N 36917 ms • 30176 ms • 12
- qwen/qwen3-14b 26323 ms
- p95 • avg • N 50867 ms • 29406 ms • 12
- qwen/qwen3-8b 25520 ms
- p95 • avg • N 32497 ms • 27019 ms • 12
- meta-llama/llama-3.1-8b… 23883 ms
- p95 • avg • N 31108 ms • 23738 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
39252992
Dec. 17, 2025, 12:01 a.m.
54900001
Dec. 16, 2025, 12:01 a.m.
34588950
Dec. 15, 2025, 12:01 a.m.
36244769
Dec. 14, 2025, 12:01 a.m.
35178382
Dec. 13, 2025, 12:01 a.m.
48156086
Dec. 12, 2025, 12:01 a.m.
44434149
Dec. 11, 2025, 12:01 a.m.
36814715
Dec. 10, 2025, 12:01 a.m.
50610334
Dec. 9, 2025, 12:01 a.m.
39104890
Dec. 8, 2025, 12:01 a.m.