NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-11-19-01M7SQ756NHDJSA7GFD82E4EGE
Run Details
Model:
Gemini 4 Neon (Reasoning, Low)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
50.3s
Cost:
$0.087186
Provider:
Google
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
96.97%
Dialogue to Total Word Ratio
(details)
98.4855%
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 tapped the red dry-erase marker against the laminated blueprint of the Hendersons' backyard. A single bare lightbulb swung lazily above the folding table, casting dramatic shadows over a detailed diagram of a plastic slide and an inflatable pirate ship. She had spent three days conducting surveillance from a parked minivan with a pair of high-powered binoculars and a box of stale donuts. "The perimeter is much tighter than we anticipated," Molly said, her voice dropping to a gravelly whisper. "They hired a professional magician." Bill looked up from his tactical vest, where he was carefully wedging a juice box into a magazine pouch. He was a man built like a refrigerator, though currently he was wearing a pastel yellow polo shirt to blend in with the suburban dads. "A magician? Like, the rabbit-out-of-a-hat kind, or the guy who sawed my cousin in half in Vegas?" Bill asked. "Worse," Molly replied, sliding a grainy photograph across the table. "Balloon animals. He has eyes everywhere. If we move on the gift table before the piƱata drops, he will spot us." Bill picked up the photo. It showed a balding man in a sequined vest twisting a long green balloon into what vaguely resembled a dachshund. Bill shuddered slightly, remembering a traumatic incident at his own sixth birthday involving a rogue clown and a bucket of confetti. "Look, Molly, are we sure the payout is worth the risk?" Bill said, gesturing vaguely at the blueprint. "We are talking about seven-year-olds. What is the haul? Socks? Plastic dinosaurs?" Molly scoffed, pacing the length of the damp garage. The Hendersons were old money, the kind of people who bought organic grapes and owned a golden retriever named Barnaby. She stopped by the whiteboard, where she had taped printouts of the birthday boy's online wish list. "Do not insult my intelligence, Bill," she snapped. "Little Jasper requested a limited-edition Millennium Falcon Lego set. Unopened, that thing flips for eight hundred bucks on the secondary market. Plus, his grandmother is flying in from Zurich. She brings cash." Bill whistled low and slow. Eight hundred dollars was enough to finally fix the transmission on their getaway sedan. He leaned over the table, suddenly invested in the logistics of the operation. His thick finger traced a path from the side gate, past the kiddie pool, and directly toward the covered patio where the loot would be stacked. "Okay, I am in," Bill murmured. "We hit the table during the cake. Nobody watches the presents when there is buttercream frosting. Just promise me we can grab a couple of goodie bags on the way out." Molly smiled, a cold, calculating expression. She uncapped her marker and drew a bold red X right over the patio. They would synchronize their watches, wait for the off-key chorus of Happy Birthday, and vanish into the hydrangeas with the plastic bricks. It was going to be the cleanest score of their careers, provided Bill did not get distracted by the bouncy castle again.
created_at
"2026-11-19T10:56:40.669Z"
11/19/2026, 10:56:40 AM
model
"google/gemini-4-neon-20261118"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1880
moderation_latency
(null)
generation_time
49992
tokens_prompt
75
tokens_completion
1791
native_tokens_prompt
97
native_tokens_completion
5793
native_tokens_completion_images
0
native_tokens_reasoning
5152
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
0.087186
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.22.2; linux; x64))"
http_referer
(null)
request_id
"req-1795085800-7mg4KDPcbehtH2oHDPzF"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1795085800-jU9zpZt8UHqanm7PkpIL"
upstream_id
"6NX-asTRLN-jq8YPlbuJ4Ag"
provider_responses
0
endpoint_id
"cbc1d01f-e431-41ec-9abb-36c1a5b10cb4"
id
"6NX-asTRLN-jq8YPlbuJ4Ag"
is_byok
false
latency
1880
model_permaslug
"google/gemini-4-neon-20261118"
provider_name
"Google"
status
200
total_cost
0.087186
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
96.97%
Dialogue to Total Word Ratio
Ratio: 30.78%, Deviation: 0.78%
neededClean
false
wordsTotal
510
wordsDialogue
157
98.4855%