NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHYB9B2TEGWBQ8D0Q8FD222K
Run Details
Model:
Gemma 3 12B
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
5.2s
Cost:
$0.000039
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
97.63%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
48.8138%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 50% 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 inflatable bouncy castle loomed, a garish pink monstrosity against the fading sunlight. Molly chewed on a licorice whip, surveying the carnage of party games and sugar-fueled children. “Right then, Operation Sugar Rush is a go,” she announced, not entirely seriously. Bill adjusted his ridiculous fake mustache. “Are we *sure* about this, Molly? Stealing presents from five-year-olds? It feels… low.” “Low is paying rent, Bill! Besides, think of the resale value on a limited-edition Sparkle Unicorn. Those things go for a fortune on eBay.” “A fortune? Really?” “Absolutely. Look, the plan is simple. While I distract the parents with a fabricated sob story about my poorly poodle, you're in charge of present procurement.” “Procurement? Sounds fancy for grabbing toys.” Bill muttered, nervously eyeing a small boy wielding a plastic lightsaber. “And what’s the sob story? Poodles aren't that sympathetic.” “I’m working on it. Something about a rare, hypoallergenic condition and a desperate need for a specialist vet. Trust me, it’ll tug at their heartstrings… and their wallets. Now, are you with me, or are you going to let a bunch of glitter-covered toddlers deprive us of our due?”
created_at
"2026-02-20T20:19:01.351Z"
2/20/2026, 8:19:01 PM
model
"google/gemma-3-12b-it"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
379
moderation_latency
(null)
generation_time
5065
tokens_prompt
93
tokens_completion
330
native_tokens_prompt
75
native_tokens_completion
279
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
0.00003927
router
(null)
provider_responses
0
endpoint_id
"eb06dc92-5a16-47ec-a776-6ef956457c47"
id
"chatcmpl-R7rgxt0AeGg2ARtHBykX6rB3"
is_byok
false
latency
61
model_permaslug
"google/gemma-3-12b-it"
provider_name
"DeepInfra"
status
200
api_type
"completions"
id
"gen-1771618741-nruBCNUjEEOuTbpCsJ6e"
upstream_id
"chatcmpl-R7rgxt0AeGg2ARtHBykX6rB3"
total_cost
0.00003927
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepInfra"
Evaluation details
Result
Evaluator
Details
Meta Data
97.63%
Matches word count
n/a
neededClean
false
words
193
0.00%
Dialogue to Total Word Ratio
Ratio: 73.58%, Deviation: 23.58%
neededClean
false
wordsTotal
193
wordsDialogue
142
48.8138%