NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-11-01KKFD2XZW52A6B44XQP5EC587
Run Details
Model:
Stealth: Hunter Alpha
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
11.2s
Cost:
$0.000000
Provider:
Stealth
Evaluation summary
Score
Evaluator
Details Link
3.62%
Matches word count
(details)
36.79%
Dialogue to Total Word Ratio
(details)
20.2057%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 90% words of dialogue between two thieves named Molly and Bill who are planning to break into a children's party to steal all the presents. Only include the prose of the scene without any markdown or commentary.
Model response
Messages
Meta
AI
“Alright, Bill. The guest of honor is a five-year-old named Timmy. He’s terrified of clowns.” “So you’re dressing as a clown?” “I’m dressing as *his* clown. The one who cancelled. The ‘Giggles’ who called in sick.” “Brilliant. I’ll be the magician.” “You can’t do magic.” “I don’t need to do magic. I just need to misdirect. Watch me. I’ll pull a quarter from Timmy’s ear, and while he’s checking for a hole, you’re already in the gift room.” “The gift room is next to the cake table. There’s a mother there, Molly. A *craft* mother. She’s got a hot glue gun and she’s not afraid to use it.” “I’ve studied the floor plan. The glitter volcano erupts at 3:15. That’s our window. The ensuing chaos will cover the sound of the wrapping paper rustling.” “What about the piñata?” “The piñata is the getaway vehicle.” “You want to stuff the presents into a papier-mâché donkey?” “It’s a *pony*, Bill. And yes. We walk out carrying the pony, full of loot. No one suspects the piñata.” “Except the kids. They’ll want to smash it.” “That’s where your ‘magic’ comes in. You promise them it’s a *magic* piñata that only breaks for the birthday boy. Tomorrow. After we’ve sold the contents.” “You’re a monster, Molly.” “I’m a professional. Now go practice pulling quarters out of thin air.”
created_at
"2026-03-11T21:33:15.655Z"
3/11/2026, 9:33:15 PM
model
"openrouter/hunter-alpha"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1475
moderation_latency
(null)
generation_time
11104
tokens_prompt
77
tokens_completion
629
native_tokens_prompt
2111
native_tokens_completion
546
native_tokens_completion_images
(null)
native_tokens_reasoning
250
native_tokens_cached
2048
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
usage
0
router
(null)
provider_responses
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.0; linux; x64))"
http_referer
(null)
api_type
"completions"
id
"gen-1773264795-mjBeQp2Okr3v3LH3nY8q"
upstream_id
"gen-1773264795-mjBeQp2Okr3v3LH3nY8q"
total_cost
0
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Stealth"
Evaluation details
Result
Evaluator
Details
Meta Data
3.62%
Matches word count
n/a
neededClean
false
words
224
36.79%
Dialogue to Total Word Ratio
Ratio: 100.00%, Deviation: 10.00%
neededClean
false
wordsTotal
227
wordsDialogue
227
20.2057%