NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHY81EX785CFBGRNJD1PF2X4
Run Details
Model:
Gemma 3 12B
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
20.2s
Cost:
$0.000090
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
68.56%
Dialogue to Total Word Ratio
(details)
34.2785%
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 laundromat hummed a monotonous tune, a soundtrack to Molly’s frantic gesturing with a crumpled napkin. Bill, perched on a folding chair, meticulously oiled a miniature grappling hook. The scent of damp cotton and cheap detergent hung heavy in the air. “So, just to recap,” Molly said, stabbing the napkin with a pink nail, “we're robbing a children's birthday party.” Bill didn’t look up. “‘Acquiring’ is a more appropriate term, Molly. We’re acquiring gifts.” “Acquiring gifts from six-year-olds,” she repeated, letting the words drip with incredulity. “Bill, I envisioned us hitting a jewelry store, maybe a small bank. Not...’Happy 6th Birthday, Penelope!’” “Penelope’s dad owns three shipping yards,” Bill mumbled, still focused on the grappling hook. “Shipping yards that, I might add, are deeply underinsured.” “That’s irrelevant! We’re dressing up as clowns to steal Lego and My Little Ponies. My reputation will be in tatters!” “Relax, Molly. Nobody suspects a clown.” Bill finally looked up, a smear of oil on his cheek. “Especially a clown with a slightly unsettling grin and a penchant for escaping from oversized shoes.” Molly glared. “My grin is charming! And the shoes are regulation. Besides, how do you even *know* what Penelope wants? We can’t just walk in and demand all the princess-themed paraphernalia. There'll be tears!” “We have intel, Molly! Ingrid at the bakery overheard Penelope’s mother bragging about a diamond-encrusted tiara.” “A diamond-encrusted tiara! For a six-year-old?” Molly’s eyes widened. “Okay, now you have my attention. But the rest of the presents? We can’t just claim tiaras.” “We'll work with the theme. Pony-based items, Lego sets geared towards imaginative play, anything with glitter. The more offensively cute, the better. It throws people off.” Bill tested the grappling hook, sending a tiny hook spinning through the air and sticking briefly to a pile of freshly folded towels. “Right. Offensively cute. And the distraction? You’re still insisting on the pigeon release?” “Essential,” Bill stated firmly. “A sudden flurry of pigeons will provide maximum chaos. Everyone will be looking up, and we'll be in and out with the loot before anyone can say ‘pin the tail on the donkey’.” “Pigeons. Seriously? You couldn’t just hire a juggler or something?” "Jugglers are predictable. Pigeons are an act of God. Pure, feathered anarchy." Bill returned to his oiling. "Plus, I got a good deal on them. Old Mrs. Higgins down the street was overrun. Said they kept stealing her dentures.” Molly groaned. “Of course she did. Fine, pigeons. Just… please tell me you’ve got the clown makeup allergy covered.” "Naturally. Contact lenses. And a strategically placed antihistamine." Bill winked. “Now, about these oversized shoes…”
created_at
"2026-02-20T19:22:17.4Z"
2/20/2026, 7:22:17 PM
model
"google/gemma-3-12b-it"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
851
moderation_latency
(null)
generation_time
20025
tokens_prompt
93
tokens_completion
769
native_tokens_prompt
75
native_tokens_completion
666
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
0.00008958
router
(null)
provider_responses
0
endpoint_id
"eb06dc92-5a16-47ec-a776-6ef956457c47"
id
"chatcmpl-RqpESr3qGdk3QL7YHibfjIjd"
is_byok
false
latency
60
model_permaslug
"google/gemma-3-12b-it"
provider_name
"DeepInfra"
status
200
api_type
"completions"
id
"gen-1771615337-IeHdaZS61K6zuWkkx0SO"
upstream_id
"chatcmpl-RqpESr3qGdk3QL7YHibfjIjd"
total_cost
0.00008958
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepInfra"
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
444
68.56%
Dialogue to Total Word Ratio
Ratio: 72.75%, Deviation: 2.75%
neededClean
false
wordsTotal
444
wordsDialogue
323
34.2785%