NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHY7HXMZZRP0Z5WGXRDY7Z8C
Run Details
Model:
Gemma 3 12B
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
9.0s
Cost:
$0.000040
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
86.38%
Matches word count
(details)
99.55%
Dialogue to Total Word Ratio
(details)
92.9648%
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
"Right, Bill, so we’re agreed? This is foolproof." Molly tapped a brightly colored diagram of a bouncy castle. "Operation Sugar Rush is a go." Bill squinted. "Foolproof? Molly, we’re robbing a children's party. They’ll be covered in glitter and wielding plastic swords. How is *that* foolproof?" "Think of the presents, Bill! Limited edition superhero figurines, remote-controlled cars, the works! A veritable treasure trove!" "A treasure trove that requires us to navigate a horde of sugar-fueled five-year-olds." Bill sighed. "What's the plan for distraction? You said you were handling that?" “Of course! I’m bringing a magician. A really bad one. He does rabbit-out-of-hat tricks using a suspiciously small hat. Guaranteed to mesmerize them for at least five minutes.” "Five minutes? Molly, those kids have the attention span of goldfish. Five minutes and they'll be chanting 'More glitter! More cake!'" "Then we move quickly! We grab the loot, disappear into the shrubbery. Simple!" "Simple involving pygmy hippos, maybe. Are we even sure there’ll be anything worth stealing?” "Trust me, Bill. There will be. Now, did you remember the grappling hook shaped like a unicorn?"
created_at
"2026-02-20T19:13:48.203Z"
2/20/2026, 7:13:48 PM
model
"google/gemma-3-12b-it"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
626
moderation_latency
(null)
generation_time
8788
tokens_prompt
93
tokens_completion
315
native_tokens_prompt
75
native_tokens_completion
287
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
0.00004031
router
(null)
provider_responses
0
endpoint_id
"eb06dc92-5a16-47ec-a776-6ef956457c47"
id
"chatcmpl-RtWRS7TBGBzm0HkfAVCdwnvI"
is_byok
false
latency
119
model_permaslug
"google/gemma-3-12b-it"
provider_name
"DeepInfra"
status
200
api_type
"completions"
id
"gen-1771614828-6ImTl3wC29fwMXBNrwUD"
upstream_id
"chatcmpl-RtWRS7TBGBzm0HkfAVCdwnvI"
total_cost
0.00004031
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepInfra"
Evaluation details
Result
Evaluator
Details
Meta Data
86.38%
Matches word count
n/a
neededClean
false
words
189
99.55%
Dialogue to Total Word Ratio
Ratio: 92.59%, Deviation: 2.59%
neededClean
false
wordsTotal
189
wordsDialogue
175
92.9648%