NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M37HAEYZ2P64VM8C6JSJFFWA
Run Details
Model:
GPT-6 Sol (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
18.0s
Cost:
$0.009192
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
14.30%
Matches word count
(details)
6.74%
Dialogue to Total Word Ratio
(details)
10.5211%
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 community hall glittered with paper stars. Through its window, Molly watched twenty children attack a piñata with the determination of tiny medieval soldiers. Beside her, Bill unfolded a floor plan across the hood of a parked ice-cream van. “You drew the kitchen twice,” Molly said. “One’s the real kitchen. The other’s where they keep the tiny sandwiches.” “That’s also the kitchen.” “Then their security is worse than I thought.” Molly pointed at a mountain of wrapped boxes inside the hall. “There’s the target. Every present leaves with us before the cake comes out.” “Why before cake?” “Because after cake, they get faster.” Bill studied the window. “I’ve brought disguises.” “Please say caterers.” “Better. Magicians.” “We don’t know any magic.” “I can make a fifty-pound note disappear.” “You mean spend it.” “The audience won’t know the difference.” A cheer erupted from the hall. The piñata had surrendered. “We go in through the side door,” Molly said. “You distract the adults. I load the presents onto the trolley.” “With what act?” “Pull a rabbit from a hat.” “We haven’t got a rabbit.” “Pull a hat from a rabbit. Improvise.” Bill scratched his chin. “What if a child asks to help?” “Give them a wand and say they’re in charge of smoke.” “That sounds dangerous.” “It’s a cardboard tube. We don’t have smoke.” “Then what’s in this?” He lifted a dented canister from the van. Molly stared at the label. “Whipped cream.” “The man at the shop said it was party supplies.” “It is. Put it back.” Inside, a woman began counting down from ten. A fresh shriek of excitement rattled the glass. “Cake?” Bill asked. “Pass the parcel,” Molly said. “We’re running out of time.” He folded the floor plan. “I’ve had a thought.” “Try not to alarm it.” “If we steal every present, the children will cry.” “That is generally how stealing works.” “Not if we replace them.” “With what?” Bill opened the van’s rear doors. Stacked to the ceiling were boxes of unsold novelty slippers: sharks, dinosaurs, and one pair shaped like a disappointed accountant. “You bought all those?” “I won them in a poker game.” “You’re terrible at poker.” “Yes, but imagine how bad the other fellow was.” Molly considered the slippers, then the presents. “We’ll swap them. Nothing under the gift table when the parents come looking.” “Now we sound like professionals.” “We are professionals. You’re just a professional mistake.” Bill grinned and reached for his magician’s cape. “Wait,” Molly said. “Why does the van say MISTER WHIPPY’S CHILDREN’S PARTIES?” “It came with the slippers.” The side door opened. A boy in a party hat peered out at them, then at the ice-cream van. “New plan,” Molly whispered. “You serve ice cream. I move the presents.” “And the magic?” “You’re about to make a hundred cones disappear.” Bill straightened his cape. “Finally. An act I’ve rehearsed.”
created_at
"2026-09-23T16:24:30.613Z"
9/23/2026, 4:24:30 PM
model
"openai/gpt-6-sol-20260922"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
5402
moderation_latency
(null)
generation_time
17697
tokens_prompt
75
tokens_completion
1091
native_tokens_prompt
66
native_tokens_completion
906
native_tokens_completion_images
0
native_tokens_reasoning
236
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
0.009192
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790180670-oBUGMZpjhYqz511efYEe"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790180670-K5grNXIRVHf2dJfg0KIx"
upstream_id
"resp_085b187571e0c97d016ab3fd3ec55487d19abaf53227a9c564"
provider_responses
0
endpoint_id
"1cfc7d9d-4404-4b8e-9ee4-58ae45c9dcd4"
id
"resp_085b187571e0c97d016ab3fd3ec55487d19abaf53227a9c564"
is_byok
false
latency
573
model_permaslug
"openai/gpt-6-sol-20260922"
provider_name
"OpenAI"
status
200
total_cost
0.009192
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
14.30%
Matches word count
n/a
neededClean
false
words
479
6.74%
Dialogue to Total Word Ratio
Ratio: 62.66%, Deviation: 7.34%
neededClean
false
wordsTotal
482
wordsDialogue
302
10.5211%