NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-03-01KJTZPGV69TKMABXDDF2RDV5X
Run Details
Model:
Gemini 3.1 Flash Lite (Preview)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
3.9s
Cost:
$0.000918
Provider:
Google AI Studio
Evaluation summary
Score
Evaluator
Details Link
35.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
17.5011%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 30% 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
Molly adjusted her neon pink wig, the synthetic curls itching against her neck like a swarm of angry ants. She leaned against the peeling wallpaper of the van, glaring at the clipboard clutched by Bill. He was currently wearing a velvet waistcoat that smelled faintly of mothballs and despair. Outside, the suburban street was silent, save for the distant, rhythmic thumping of a bass-heavy children’s song vibrating through the pavement. The target was a sprawling ranch house decorated with enough balloons to lift a toddler into low earth orbit. Everything is set, Molly said, tapping her finger against her fake plastic nose. The clown show starts in twenty minutes. Are you sure about the clown? Bill sighed, checking his pocket watch for the third time in a minute. The clown is an independent contractor, Molly. He has a history of professionalism and a deep hatred for late arrivals. He will be there, and he will provide the perfect distraction while we secure the loot. Molly checked the lock-picking set tucked into her polka-dot belt. I still think we could have just crawled through the doggy door. It might have been more dignified than pretending to be the entertainment duo known as Giggles and Groan. Dignity is for people with pensions, Bill replied, smoothing his oversized bowtie. We are here for the loot, and the loot is currently sitting under a mountain of cheap wrapping paper in the living room. Imagine the haul: Lego sets, tablets, whatever luxury toys those spoiled children are hoarding this year. It will be our greatest score. Molly rolled her eyes, though she couldn't suppress a grin. You better hope there is a cake. I heard the lady of the house spent a fortune on one of those custom sugar-sculpture monstrosities. If we walk away with the presents but no cake, I am going to be very disappointed. Bill shoved the clipboard into his bag and patted her shoulder. We are thieves, not gluttons. Focus on the mission. Once the clown starts his balloon animal routine, we drop down from the skylight. We navigate the gift corridor, bag the haul, and exit through the laundry room. It is a foolproof plan, provided you do not get distracted by the petting zoo in the backyard again. That goat had a vendetta, Molly insisted, peering through the binoculars. It looked at me with the eyes of a hardened criminal. Besides, the mission is simple. We go in, we grab the wrapping paper bounty, and we vanish before the parents realize their precious offspring are getting less than their money's worth. Bill opened the door, a wave of sugary music flooding the van. We leave in sixty seconds. Try to remember that you are supposed to be funny, not terrifying. Molly winked, adjusting her giant clown shoes. I will be hilarious, Bill. Just try to keep up with the gravity-defying acrobatics.
created_at
"2026-03-03T23:14:28.85Z"
3/3/2026, 11:14:28 PM
model
"google/gemini-3.1-flash-lite-preview-20260303"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
556
moderation_latency
(null)
generation_time
3771
tokens_prompt
77
tokens_completion
721
native_tokens_prompt
67
native_tokens_completion
601
native_tokens_completion_images
0
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
0.00091825
router
(null)
provider_responses
0
endpoint_id
"e004ce39-eb0c-4474-93d5-55072e725970"
is_byok
false
latency
556
model_permaslug
"google/gemini-3.1-flash-lite-preview-20260303"
provider_name
"Google AI Studio"
status
200
api_type
"completions"
id
"gen-1772579668-wUmEYaU5xbW5Bd39N6Mk"
upstream_id
(null)
total_cost
0.00091825
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Google AI Studio"
Evaluation details
Result
Evaluator
Details
Meta Data
35.00%
Matches word count
n/a
neededClean
false
words
482
0.00%
Dialogue to Total Word Ratio
Ratio: 0.00%, Deviation: 30.00%
neededClean
false
wordsTotal
487
wordsDialogue
0
17.5011%