NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-03-01KJV0BW68PAWZQG4Y74FFF3BP
Run Details
Model:
Gemini 3.1 Flash Lite (Preview)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
3.5s
Cost:
$0.000960
Provider:
Google AI Studio
Evaluation summary
Score
Evaluator
Details Link
43.38%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
21.6892%
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 living room ceiling fan hummed, barely cutting through the thick, suburban heat. Molly tapped her pen against a blueprint of the Miller residence, her brow furrowed in mock concentration. Bill sat opposite her, his fingers drumming a nervous beat against a half-eaten sandwich. Are you absolutely certain about the clown, Bill? Because if he catches us, I am not sharing the loot. He is not a clown, Molly. He is a balloon artist named Mr. Giggles. And he is seventy-two years old and mostly blind. Mostly blind is still dangerous. What if he accidentally pops a balloon in my face while I am trying to liberate the Legos? You need to focus. The perimeter is clear. The parents will be inside, distracted by a puppet show. That is our window. And what about the dog? You said nothing about a dog. It is a Golden Retriever, Molly. It would probably help us carry the presents to the car. I refuse to bribe a dog. Do you have the grappling hook? Why would I need a grappling hook for a ranch-style house? There is a sliding glass door. It is locked, obviously, but a sturdy butter knife will handle the latch. You are so unimaginative. A grappling hook adds flair. It suggests a certain level of professional prestige. It suggests we are trying to scale a ten-foot wall that does not exist. Did you bring the pillowcases, or are we just going to carry the gifts out like confused delivery people? I have the pillowcases. The superhero ones. They were on sale. Perfect. We will blend right in. The kids will just think we are the world's most aggressive gift-givers. I am still worried about the cake. If we move too fast, we might trigger the confetti cannon. There is a confetti cannon? It is a seven-year-old’s birthday party, Bill. There is always a confetti cannon. Just move quietly. If we grab the presents near the tree first, we can exit through the kitchen before the sugar rush hits. And if we get caught? Then you tell them you are the entertainment. You have always been a fantastic mime. I hate you. I know. Now tuck your shirt in. We leave in ten minutes. Do I have to wear the oversized clown shoes again? They are for stealth, Molly. Rubber soles make for silent footsteps. They have bells on the toes, Bill. Exactly. It confuses the enemy. They will think they are hearing wind chimes, not a master thief. You are a lunatic. We are going to prison because of your obsession with rhythmic bells. Prison is just a place with fewer birthdays. Grab your crowbar. It is showtime. Fine, but if I end up in a holding cell wearing balloon-animal ears, I am naming you as the accomplice immediately. Deal. Now, check the perimeter. I think Giggles just started inflating a poodle.
created_at
"2026-03-03T23:26:08.597Z"
3/3/2026, 11:26:08 PM
model
"google/gemini-3.1-flash-lite-preview-20260303"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
663
moderation_latency
(null)
generation_time
3370
tokens_prompt
77
tokens_completion
723
native_tokens_prompt
67
native_tokens_completion
629
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
0.00096025
router
(null)
provider_responses
0
endpoint_id
"e004ce39-eb0c-4474-93d5-55072e725970"
is_byok
false
latency
663
model_permaslug
"google/gemini-3.1-flash-lite-preview-20260303"
provider_name
"Google AI Studio"
status
200
api_type
"completions"
id
"gen-1772580368-ahhJKUq5I5jPtR1cA1O6"
upstream_id
(null)
total_cost
0.00096025
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Google AI Studio"
Evaluation details
Result
Evaluator
Details
Meta Data
43.38%
Matches word count
n/a
neededClean
false
words
483
0.00%
Dialogue to Total Word Ratio
Ratio: 0.00%, Deviation: 70.00%
neededClean
false
wordsTotal
491
wordsDialogue
0
21.6892%