NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-19-01KHVPF0FW3C9B3VEVRRWJAN53
Run Details
Model:
Gemini 3.1 Pro (Preview)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
3m 38s
Cost:
$0.215474
Provider:
Google AI Studio
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
100.00%
Dialogue to Total Word Ratio
(details)
100.0000%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 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 primary target is the bounce house." Molly patiently tapped her plastic pen against the heavily smeared blueprints taped to the dashboard. "The host piled the loot right beside the inflatable moat." "What about the hired clowns?" Bill adjusted the strap of his tactical night vision goggles and squinted at the sunlit driveway across the street. "Clowns are highly unpredictable variables. They have the squeaky shoes. They possess mathematically impossible handkerchiefs." "Forget the clowns. We go in during the pinata distraction." Molly reached down and pulled a tiny juice box from the red cooler. "Once the cardboard donkey bursts, we strike. We cannot miss this window of opportunity." "Pinatas cause mass hysteria." Bill slowly tapped a thick index finger directly against his deeply calloused chin. "Are we anticipating sugar squads? Because a seven year old hopped up on funfetti cake has the exact same combat awareness as an angry honey badger." Molly sighed heavily, watching a golden retriever happily chase a frisbee across the manicured lawn of their suburban target. She rolled down the window of the surveillance van to let in some air. The faint smell of hot dogs wafted over. "There will be cake, yes. But we bypass the patio entirely. You will enter through the side gate disguised as the hired petting zoo handler." Molly sighed and ruthlessly handed him a fluffy pair of aggressively pink novelty rabbit ears. Bill stared at the fleece accessory as if it were a live grenade. "I specifically refuse to wear the fuzzy ears. The ears severely compromise my peripheral vision in the field." Molly leaned forward and mercilessly forced the ridiculous costume ears firmly onto his bald head. "They sell the illusion. Now, remember the extraction protocol. We grab the large rectangular packages first. That is where the premium electronics live. Leave the soft bags." "Soft bags are always clothes. Am I authorized to engage the hostiles if they offer me a juice pouch?" Bill nervously tested the tight elasticity of his chin strap, openly wincing as the ears leaned left. Molly verified her printed invitation font against the forged credentials again. She slid a sleek crowbar into a floral tote bag along with standard issue zip ties and a deeply suspicious looking casserole dish meant to serve as their primary cover story during the operation. "No engaging. Even eye contact with a toddler could trigger a tantrum alarm. If a mother approaches you, speak exclusively in farm animal noises until she retreats." Molly violently clicked her tongue in sheer and absolute frustration. Bill nodded very sharply, fiercely committing the incredibly complex tactical sequence to his precise eidetic memory. "So I grab the premium gifts, avoid the clown, ignore the sugar badgers, and bleat like a goat. And what exactly are you doing during all this?" Molly actually smiled broadly, finally grabbing her heavy black tactical binoculars from underneath the broken passenger seat cushion. "I will be sitting right at the craft table. Someone needs to supervise the macaroni necklaces."
created_at
"2026-02-19T19:36:37.976Z"
2/19/2026, 7:36:37 PM
model
"google/gemini-3.1-pro-preview-20260219"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
5021
moderation_latency
(null)
generation_time
217787
tokens_prompt
77
tokens_completion
5824
native_tokens_prompt
67
native_tokens_completion
17945
native_tokens_completion_images
0
native_tokens_reasoning
17333
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
0.215474
router
(null)
provider_responses
0
endpoint_id
"2abf5677-0826-45d9-8470-3d62c31d54a1"
is_byok
false
latency
5021
model_permaslug
"google/gemini-3.1-pro-preview-20260219"
provider_name
"Google AI Studio"
status
200
api_type
"completions"
id
"gen-1771529797-CDFhbcMOwxTDInMSRPlo"
upstream_id
(null)
total_cost
0.215474
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Google AI Studio"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
500
100.00%
Dialogue to Total Word Ratio
Ratio: 50.00%, Deviation: 0.00%
neededClean
false
wordsTotal
500
wordsDialogue
250
100.0000%