Miles West
food-hospitality-culinary-arts-fast-food-shop-owner-characters-harland-david-sanders
v2.0
Ethical
Backstory: A self-made entrepreneur, Miles began with a single food cart after culinary school and now runs a small chain of gourmet-inspired burger shops. He champions fresh, locally sourced ingredients, rotates regional specials weekly, and greets regulars by name. Passionate about hospitality, he trains staff to pair warmth with modern ordering tech and eco-friendly packaging.
100% Complete
4/4 scenes
Model Performance Overview
Scene Performance Matrix
| Scene | deepseek/deepseek-r… | google/gemini-2.5-f… | google/gemma-3-12b-… | meta-llama/llama-3.… | microsoft/phi-3-med… | microsoft/phi-3.5-m… | mistralai/mistral-7… | neversleep/noromaid… | [email protected]… | [email protected]… | qwen/qwen-2.5-7b-in… | qwen/qwen3-14b | qwen/qwen3-8b |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
first-time-guest
Greeting a New Customer
|
0.596
Details |
0.672
Details |
0.682
Details |
0.557
Details |
0.000
Details
Error
|
0.627
Details |
0.682
Details |
0.621
Details |
0.000
Details
Error
|
0.665
Details |
0.575
Details |
0.693
Details |
0.688
Details |
loyal-customer
Remembering a Regular
|
0.740
Details |
0.863
Details |
0.730
Details |
0.628
Details |
0.000
Details
Error
|
0.833
Details |
0.759
Details |
0.339
Details |
0.000
Details
Error
|
0.737
Details |
0.728
Details |
0.524
Details |
0.715
Details |
weekly-menu-post
Instagram Menu Announcement
|
0.000
Details |
0.401
Details |
0.406
Details |
0.366
Details |
0.000
Details
Error
|
0.000
Details
Error
|
0.354
Details |
0.000
Details |
0.000
Details
Error
|
0.541
Details |
0.233
Details |
0.251
Details |
0.483
Details |
staff-training-module
Onboarding Guide Snippet
|
0.386
Details |
0.547
Details |
0.519
Details |
0.000
Details |
0.000
Details |
0.548
Details |
0.252
Details |
0.000
Details |
0.000
Details
Error
|
0.628
Details |
0.334
Details |
0.663
Details |
0.450
Details |
Test Scenes 4
0
Scene Order
Greeting a New Customer
ID:
first-time-guest
🎯 Goal:
Offer a warm, concise welcome (≤100 words), spotlight fresh local sourcing, and recommend today’s regional special.
📨 Input Events:
chat_msg
viewer:new_customer
"Hi, it's my first time here. What's good?"
Ready for Testing
1
Scene Order
Remembering a Regular
ID:
loyal-customer
🎯 Goal:
Greet the customer by name, recall their usual order, and suggest a small variation, all in a friendly, personal tone (≤120 words).
🧠 Initial State:
Pre-loaded Memories:
- 💭 {'kind': 'fact', 'tags': ['customer_profile'], 'content': 'Regular customer Jada normally orders a Blue Ridge BBQ Burger with extra pickles.', 'importance': 2}
📨 Input Events:
chat_msg
viewer:loyal_customer
"Hey Miles, back again for my usual!"
Ready for Testing
2
Scene Order
Instagram Menu Announcement
ID:
weekly-menu-post
🎯 Goal:
Generate a 150–200 word Instagram post unveiling next week’s rotating regional burger specials, emphasizing local produce and compostable packaging, with an upbeat voice and one branded hashtag.
📨 Input Events:
world_event
system
"It's Sunday evening; prepare next week's Instagram announcement."
Ready for Testing
3
Scene Order
Onboarding Guide Snippet
ID:
staff-training-module
🎯 Goal:
Write a 250–300 word training snippet for new staff covering greeting scripts, mobile ordering tablets, and recycling/compost procedures, using a warm coaching tone.
📨 Input Events:
chat_msg
assistant_manager
"Could you draft the onboarding notes for our new hires?"
Ready for Testing
Latency by Model (This Suite)
Fastest
- [email protected]/Qw… 12252 ms
- p95 • avg • N 14806 ms • 12517 ms • 4
- google/gemini-2.5-flash 17920 ms
- p95 • avg • N 20296 ms • 17915 ms • 4
- google/gemma-3-12b-it 19929 ms
- p95 • avg • N 21035 ms • 19776 ms • 4
- meta-llama/llama-3.1-8b… 20500 ms
- p95 • avg • N 35165 ms • 22508 ms • 4
- qwen/qwen-2.5-7b-instru… 21459 ms
- p95 • avg • N 33273 ms • 24537 ms • 4
Slowest
- microsoft/phi-3-medium-… 111662 ms
- p95 • avg • N 127453 ms • 114546 ms • 4
- [email protected]/Qw… 41007 ms
- p95 • avg • N 46078 ms • 41653 ms • 4
- microsoft/phi-3.5-mini-… 38052 ms
- p95 • avg • N 126466 ms • 61529 ms • 4
- neversleep/noromaid-20b 32649 ms
- p95 • avg • N 42973 ms • 32139 ms • 4
- deepseek/deepseek-r1-di… 29327 ms
- p95 • avg • N 35544 ms • 30136 ms • 4
Per-scene duration for this suite.
Suite Actions
Completion Progress
100%
4 of 4 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
23562731
Dec. 17, 2025, midnight
27873436
Dec. 16, 2025, midnight
22330331
Dec. 15, 2025, midnight
25435764
Dec. 14, 2025, midnight
22280581
Dec. 13, 2025, midnight
27311462
Dec. 12, 2025, midnight
23183672
Dec. 11, 2025, midnight
22562459
Dec. 10, 2025, midnight
25979111
Dec. 9, 2025, midnight
22903549
Dec. 8, 2025, midnight