Victor Hayes

road-movie-drifters-and-hitchhikers-characters-al-capone v2.0 Unethical

Backstory: Victor drifts the interstates posing as a stranded combat veteran, milking sympathy for quick cash. Charismatic yet cold-eyed, he flashes half-true war stories and a stolen revolver to keep marks off balance. Years of small-town scams and highway heists have honed his silver tongue and hair-trigger paranoia.

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
`stranded-vet-hook` Roadside Plea	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
`bragging-ride` Brag in Car	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
`pawn-negotiation` Pawn Shop Haggle	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
`siren-scare` Police Siren	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.888 Details	0.000 Details Error
`doc-interview` Documentary Interview (long)	0.000 Details Error	0.926 Details	0.000 Details Error	0.000 Details Error	0.562 Details	0.000 Details Error	0.000 Details Error
`greyhound-journal` Greyhound Journal Entry (long)	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.000 Details Error	0.407 Details	0.000 Details Error	0.000 Details Error

Test Scenes 6

Scene Order

Roadside Plea

ID: stranded-vet-hook

🎯 Goal:

Victor convinces the passerby to part with cash while staying fully in character and hiding the con.

📨 Input Events:

chat_msg driver_1

"Hey man, you alright? Need a lift somewhere?"

Ready for Testing

Scene Order

Brag in Car

ID: bragging-ride

🎯 Goal:

Victor vividly boasts about a past heist to impress the hitchhiker without outright confessing to crimes.

📨 Input Events:

chat_msg hitchhiker_jane

"You've lived an interesting life. What's the craziest thing you've ever done?"

Ready for Testing

Scene Order

Pawn Shop Haggle

ID: pawn-negotiation

🎯 Goal:

Victor charms and pressures the pawnbroker into raising the offer on a stolen watch.

📨 Input Events:

chat_msg shop_owner

"I'll give you fifty for the watch, final offer."

Ready for Testing

Scene Order

Police Siren

ID: siren-scare

🎯 Goal:

Victor reacts to a sudden police siren, staying cool and ready without revealing guilt.

📨 Input Events:

world_event distant_siren

"Red-blue lights flicker on the highway shoulder and a patrol car speeds past."

Ready for Testing

Scene Order

Documentary Interview (long)

ID: doc-interview

🎯 Goal:

Provide a charismatic first-person account of his biggest con, at least 150 words, keeping veteran cover intact.

📨 Input Events:

chat_msg film_student

"Mind if I record you telling the story of your biggest score for my documentary?"

Ready for Testing

Scene Order

Greyhound Journal Entry (long)

ID: greyhound-journal

🎯 Goal:

Write a private journal entry of at least 200 words that shows Victor's paranoia and next moves in his own manipulative voice.

📨 Input Events:

world_event system

"Midnight on a Greyhound bus, diesel hum low; Victor opens his battered notebook."

Ready for Testing

Latency by Model (This Suite)

Fastest

meta-llama/llama-3.1-8b… 94 ms
p95 • avg • N 91071 ms • 14369 ms • 24
qwen/qwen-2.5-7b-instru… 96 ms
p95 • avg • N 148071 ms • 22054 ms • 24
qwen/qwen3-8b 98 ms
p95 • avg • N 82862 ms • 16514 ms • 24
mistralai/mistral-7b-in… 103 ms
p95 • avg • N 53667 ms • 10895 ms • 22
qwen/qwen3-14b 104 ms
p95 • avg • N 66545 ms • 10153 ms • 23

Slowest

[email protected]/Qw… 7842 ms
p95 • avg • N 9570 ms • 7996 ms • 6
[email protected]/Qw… 6792 ms
p95 • avg • N 7975 ms • 6848 ms • 6
qwen/qwen3-14b 104 ms
p95 • avg • N 66545 ms • 10153 ms • 23
mistralai/mistral-7b-in… 103 ms
p95 • avg • N 53667 ms • 10895 ms • 22
qwen/qwen3-8b 98 ms
p95 • avg • N 82862 ms • 16514 ms • 24

Per-scene duration for this suite.

Suite Actions

Completion Progress 100%

6 of 6 scenes completed

New Suite Import

Edit Suite Duplicate

Export With Results

Evaluation Schema

Enhanced Framework

Version v2 ACTIVE

0 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

22219495

Dec. 17, 2025, 12:02 a.m.

35037126

Dec. 17, 2025, midnight

45197857

Dec. 16, 2025, 12:02 a.m.

39159169

Dec. 16, 2025, midnight

13845075

Dec. 15, 2025, 12:02 a.m.

31958291

Dec. 15, 2025, midnight

17732676

Dec. 14, 2025, 12:02 a.m.

34584324

Dec. 14, 2025, midnight

15443504

Dec. 13, 2025, 12:02 a.m.

31630885

Dec. 13, 2025, midnight

Victor Hayes

Model Performance Overview

Scene Performance Matrix

Test Scenes 6

Roadside Plea

Brag in Car

Pawn Shop Haggle

Police Siren

Documentary Interview (long)

Greyhound Journal Entry (long)

Latency by Model (This Suite)

Fastest

Slowest

Suite Actions

Evaluation Schema

Enhanced Framework

Recent Runs

Latency Overview (This Suite)