NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-11-19-01M7SNJJQR3NGFR46E9DE9MRNR
Run Details
Model:
Gemini 4 Neon (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
1m 8s
Cost:
$0.120306
Provider:
Google
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
99.62%
Dialogue to Total Word Ratio
(details)
99.8105%
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
Molly tapped a greasy finger against the crayon blueprint spread across the cramped diner booth. She had drawn it on the back of a placemat, using red for the laser-like gazes of the parents and blue for the escape route. "Listen closely, Bill. We have a four-minute window." Bill frowned, stirring his lukewarm coffee with a plastic straw. "Four minutes? Molly, there are thirty boxes on that table. Even dual-wielding duffel bags, I cannot clear a mountain of toys in four minutes." "This is not cheap plastic, you idiot. Little Barnaby's father invented an algorithm that sells ads to dogs. We are talking vintage holographic cards, custom drones, and one miniature electric Porsche." Bill whistled softly, nearly knocking over the salt shaker in his excitement. "An electric Porsche? Do you think it has leather seats?" "Focus! At three sharp, the mother brings out the artisanal gluten-free cake. Every kid will stampede toward the patio. That is when we breach the living room." "What about the security detail? You said they hired muscle." Molly lowered her voice and glanced suspiciously at their elderly waitress, who was busy refilling a nearby napkin dispenser. "They hired a magician named the Great Zucchini. My intel says he was a Navy Seal before pivoting to balloon animals. He has eyes like a hawk." "So how do we neutralize a tactical magician?" "You distract him. Wear the catering uniform, walk up, and ask for a dachshund. When he starts twisting, claim it looks like a poodle. His ego will take over for ten minutes." "That is brilliant. Wait. What about the bouncy castle?" Molly sighed deeply, rubbing her temples as she traced a jagged green square on the placemat. "It sits beneath the second-story window. If things go sideways, we jump out, bounce off the castle, and clear the hedge into our van." "Molly, I weigh two hundred pounds. If I jump onto a bouncy castle, I am not clearing a hedge. I am launching three toddlers into low Earth orbit." "Then stick to the plan! I grab the small electronics. You grab the boxed sets and the tiny car. Do not touch the piƱata." "Why not? Rich people put full-sized candy bars in there. Maybe loose cash." "Because last time we hit a bar mitzvah, you got greedy at the chocolate fountain and barely escaped the rabbi." Bill slumped back against the vinyl seat, looking genuinely wounded by the accusation. "That was a dark chocolate cascade, Molly. I am only human." "You are a professional thief. Act like it. Are you ready to ruin a seven-year-old's afternoon?" Bill grinned widely, pulling a black ski mask from his jacket pocket and resting it beside his basket of cold fries. "Born ready. Just promise we can stop for ice cream after we fence the Porsche." Molly folded the placemat neatly and slid out of the booth, dropping a crumpled five-dollar bill onto the table. "If we pull this off, Bill, I will buy you the whole parlor."
created_at
"2026-11-19T10:27:57.984Z"
11/19/2026, 10:27:57 AM
model
"google/gemini-4-neon-20261118"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1955
moderation_latency
(null)
generation_time
67371
tokens_prompt
75
tokens_completion
2356
native_tokens_prompt
97
native_tokens_completion
8001
native_tokens_completion_images
0
native_tokens_reasoning
7334
native_tokens_cached
0
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
num_fetches
(null)
web_search_engine
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"STOP"
service_tier
"default"
usage
0.120306
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.22.2; linux; x64))"
http_referer
(null)
request_id
"req-1795084078-DcpAeE9K4klnFLxMgRSj"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1795084078-Z4hYjgXLXsCVkTdcqv6y"
upstream_id
"Ls_-at62CJuG9LsP8-zryQo"
provider_responses
0
endpoint_id
"cbc1d01f-e431-41ec-9abb-36c1a5b10cb4"
id
"Ls_-at62CJuG9LsP8-zryQo"
is_byok
false
latency
1955
model_permaslug
"google/gemini-4-neon-20261118"
provider_name
"Google"
status
200
total_cost
0.120306
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Google"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
500
99.62%
Dialogue to Total Word Ratio
Ratio: 70.28%, Deviation: 0.28%
neededClean
false
wordsTotal
508
wordsDialogue
357
99.8105%