NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHY9N22X5PXFA789YM9X85M2
Run Details
Model:
Gemma 3 12B
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
18.7s
Cost:
$0.000095
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
0.21%
Matches word count
(details)
5.82%
Dialogue to Total Word Ratio
(details)
3.0156%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 70% 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
The aroma of stale pizza and desperation hung heavy in Molly’s cramped apartment. Bill was meticulously sharpening a rubber chicken with a paring knife. “A rubber chicken, Bill? Really?” Molly asked, pinching the bridge of her nose. “We’re robbing a children's party, not performing a bizarre poultry-themed sketch.” Bill didn’t look up. “Distraction, Molly. Essential. Think about it. Screaming kids, a rogue rubber chicken... chaos. It’ll be like pulling teeth from a sugar-fueled badger.” “It’ll be like being arrested by very disappointed parents,” Molly countered, gesturing at the board covered in blurry photos of bouncy castles and cake-laden tables. “We’re stealing presents, Bill. Not investigating a poultry conspiracy.” “Details, details. Every good heist needs a flourish. Besides," he added with a glint in his eye, "imagine the headlines: 'Chicken Causes Mayhem, Presents Vanished!'” Molly sighed. “You’re going to get us caught.” “Relax. I’ve been practicing the chicken squawk. Listen to this.” Bill puffed out his cheeks and emitted a truly alarming rubber chicken impersonation. Molly winced. “Please, stop. My ears are bleeding. So, logistics. Little Timmy’s tenth birthday, suburban cul-de-sac, three security guards according to Brenda’s intel. Brenda always exaggerates. She said one of them had a trained ferret.” “A ferret? Excellent. Ferrets are easily distracted. A well-placed trail of sausages…” Bill trailed off, lost in thought. “No sausages, Bill! We’re trying to be stealthy. The plan is simple: you create a diversion—preferably something less fowl—while I slip in and grab the presents. Pre-sorted, I hope. I’m not sifting through piles of princess tiaras.” Bill finally looked up, holding the sharpened rubber chicken aloft. “So, no chicken?” “Absolutely not. Think… a rogue clown. A balloon animal stampede.” “A stampede of balloon animals? Molly, that's utterly ridiculous.” “Is it more ridiculous than a man sharpening a rubber chicken to distract children?” Bill paused, considering. “Fair point. Okay, no chicken. Balloon animals it is. I’ve always wanted to work with balloons.” He frowned. “Though, where am I going to get a herd of balloons?” “Don’t ask me. You’re the creative one. And remember, no glitter bombs. Brenda said the mother is brutally sensitive to glitter. Apparently, she had a traumatic experience with a craft project in third grade.” “Right, no glitter. Got it. So, presents. What are we looking at here? Mostly toys, I presume? Remote-controlled cars, dolls, that sort of thing?” “Seems so. Brenda was helpful, for once. She even got a detailed list of gift preferences. Apparently, little Bethany desperately wants a pony… which they are *not* getting. So, stealing that would be particularly satisfying.” Bill grinned. “A pony? Now you're talking. Let's get rich, Molly. And maybe buy ourselves a small, slightly disgruntled pony.” Molly just rubbed her temples. “Just focus on the presents, Bill. And put the rubber chicken down.”
created_at
"2026-02-20T19:50:28.203Z"
2/20/2026, 7:50:28 PM
model
"google/gemma-3-12b-it"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
440
moderation_latency
(null)
generation_time
18027
tokens_prompt
93
tokens_completion
833
native_tokens_prompt
75
native_tokens_completion
711
native_tokens_completion_images
(null)
native_tokens_reasoning
0
native_tokens_cached
0
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
0.00009543
router
(null)
provider_responses
0
endpoint_id
"5aef2a85-3d1c-4659-a522-98bbb5272c6a"
is_byok
false
latency
57
model_permaslug
"google/gemma-3-12b-it"
provider_name
"Crusoe"
status
429
1
endpoint_id
"eb06dc92-5a16-47ec-a776-6ef956457c47"
id
"chatcmpl-RIgROme3GW1EEDiwTfyHYYxe"
is_byok
false
latency
58
model_permaslug
"google/gemma-3-12b-it"
provider_name
"DeepInfra"
status
200
api_type
"completions"
id
"gen-1771617028-4Tuii6KW6a4XXgkOPfwH"
upstream_id
"chatcmpl-RIgROme3GW1EEDiwTfyHYYxe"
total_cost
0.00009543
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepInfra"
Evaluation details
Result
Evaluator
Details
Meta Data
0.21%
Matches word count
n/a
neededClean
false
words
472
5.82%
Dialogue to Total Word Ratio
Ratio: 77.54%, Deviation: 7.54%
neededClean
false
wordsTotal
472
wordsDialogue
366
3.0156%