NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M37HRDG6TQR6KSRW8BTBW05A
Run Details
Model:
GPT-6 Sol (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
20.1s
Cost:
$0.008752
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
90.48%
Matches word count
(details)
8.92%
Dialogue to Total Word Ratio
(details)
49.7000%
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 spread a blueprint across the diner table. It was drawn in purple crayon, which was what the library kept next to the photocopier. A coffee ring obscured the room marked PRESENTS, but she knew its position by heart. “Four o’clock,” she said. “The magician starts. By four-ten, every adult is watching him saw a cardboard rabbit in half.” “Is the rabbit all right?” Bill asked. “It’s cardboard.” “That’s not what I asked.” Across the street, Happy Hippo Party Palace squatted beneath a mural of a grinning hippopotamus holding balloons. Its front door was guarded by a teenager in a paper crown. Molly had spent three afternoons watching the place from the diner. The teenager inspected invitations, confiscated outside snacks, and seemed willing to forgive almost anything if offered a cupcake. The presents were another matter. Staff stacked them in a storage room until cake was served. At five, someone wheeled them out on a trolley shaped like a castle. Molly had counted forty-six boxes delivered for today’s party, including one so large Bill insisted it contained a bicycle. Molly hoped he was right; his share of the haul depended on it. “We enter through the kitchen,” she said. “I wear a caterer’s apron. You carry the equipment.” “What equipment?” “The trolley.” “We don’t own a trolley.” “We will when we leave the kitchen.” Bill leaned back. He wore a fake moustache that had begun to peel away at one end. Molly had told him to save it for the job, but he said he needed practice being someone with a moustache. So far that person appeared to be deeply suspicious of pancakes. A waitress approached with the bill. Molly slid the blueprint under her plate. Bill smiled, moustache lifting at the corner, and asked for more syrup. Once the waitress left, Molly retrieved the drawing. A smear of jam now marked the kitchen entrance. “We’ll need a distraction,” Bill said. “I can trip the fire alarm.” “No alarms. They’d evacuate and take their gifts.” “I could volunteer for a magic trick. Then get stuck in the box.” “You getting stuck in a box isn’t a distraction. It’s a normal Saturday.” Bill considered this while prying his moustache loose. Outside, two vans arrived at the palace, followed by a woman carrying a cake with six candles. Molly recognized her from yesterday: the birthday child’s mother. She directed the drivers to unload a mound of folding chairs. The party was going to be bigger than Molly had expected. Then Bill sat forward so fast he knocked over the syrup. “I know,” he said. “We go in dressed as presents.” Molly watched syrup spread across the blueprint, engulfing the storage room. She pictured Bill under a ribbon, trying to fit through a kitchen door without bending his knees. “We’d need very large boxes,” she said. Bill glanced across the street. One driver had left a stack of empty chair cartons beside the vans. Molly folded the ruined blueprint, paid for breakfast, and took the last clean napkin to sketch a new route.
created_at
"2026-09-23T16:32:07.695Z"
9/23/2026, 4:32:07 PM
model
"openai/gpt-6-sol-20260922"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
5875
moderation_latency
(null)
generation_time
20099
tokens_prompt
75
tokens_completion
967
native_tokens_prompt
66
native_tokens_completion
862
native_tokens_completion_images
0
native_tokens_reasoning
221
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
"completed"
service_tier
"default"
usage
0.008752
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790181127-qkC99FDCKscq3M8nrwuK"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790181127-MQ3CiMxAQe3CuVJTTuu4"
upstream_id
"resp_02aee2e750a50e90016ab3ff07d2c487d18a73c2dc0101adbf"
provider_responses
0
endpoint_id
"1cfc7d9d-4404-4b8e-9ee4-58ae45c9dcd4"
id
"resp_02aee2e750a50e90016ab3ff07d2c487d18a73c2dc0101adbf"
is_byok
false
latency
553
model_permaslug
"openai/gpt-6-sol-20260922"
provider_name
"OpenAI"
status
200
total_cost
0.008752
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
90.48%
Matches word count
n/a
neededClean
false
words
510
8.92%
Dialogue to Total Word Ratio
Ratio: 23.05%, Deviation: 6.95%
neededClean
false
wordsTotal
512
wordsDialogue
118
49.7000%