NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M37HBRBMHG2TK5H104X36N0F
Run Details
Model:
GPT-6 Sol
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
13.8s
Cost:
$0.006782
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
68.10%
Matches word count
(details)
7.48%
Dialogue to Total Word Ratio
(details)
37.7902%
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
Molly spread a crayon drawing across the café table. It showed a house, a garden, and a purple creature with seventeen legs. “Is that the party venue?” Bill asked. “Yes.” “What’s the creature?” “Unclear. Possibly a dog. Possibly security.” Bill studied the drawing. “Where did you get this?” “From the birthday invitation in my nephew’s backpack. He drew the map himself.” “Your nephew is going to be there?” “He’s turning six.” Bill set down his biscuit. “You want to steal your nephew’s presents?” “I want to retrieve them. There’s a difference.” “He hasn’t even opened them.” “Exactly. Minimal sentimental attachment.” Outside, a balloon bobbed past the café window, tied to a woman’s wrist. Molly watched it go with the grave attention of a general observing enemy aircraft. Bill watched Molly. “Remind me why we’re doing this.” “Because my sister said I couldn’t organise a successful children’s party.” “So you’re proving her wrong by ruining one?” “I’m proving I could organise a successful heist. It’s adjacent.” Bill turned the drawing sideways. A red arrow crossed the garden and ended at a square labelled CAEK. Beside it, someone had written NO BILL in large green capitals. “Why does the map say ‘No Bill’?” Molly snatched it back. “That says ‘no ball.’ There’s a pond.” “It’s spelled with an I.” “He’s six.” “He knows me.” “He knows you as Uncle Bill, who taught him to juggle eggs.” “I taught him to hold one egg.” “And look what happened.” Molly folded the map and produced two party hats. One was silver; the other bore a smiling dinosaur. “Disguises,” she said. Bill put on the dinosaur hat. It perched above his ears like a tiny, cheerful roof. “Nobody will recognise me.” “You’ll be a parent.” “Of which child?” “Whichever one isn’t looking.” They left the café and reached the party ten minutes later. Through the open garden gate came shrieks, music, and the unmistakable sound of a grown man losing an argument with a bubble machine. Presents were stacked on a table beside the cake, each wrapped in bright paper. Molly stopped at the gate. Her nephew spotted her immediately. “Aunt Molly!” he shouted. “You came!” She glanced at Bill. “We’re compromised.” The boy ran over and hugged her knees. Behind him, Molly’s sister emerged carrying a tray of sandwiches. “I thought you couldn’t make it,” her sister said. “We had a scheduling breakthrough.” “Wonderful. Can you help with the presents? The table’s collapsing.” Bill looked at the sagging legs. “That wasn’t in the plan.” For the next hour, they carried every present inside, arranged them by size, and kept them safe from a sticky child in a cape. Molly’s nephew supervised, issuing orders through a cardboard tube. At cake time, her sister handed Molly a slice. “Couldn’t have managed without you.” Molly stared at it. Bill leaned close. “Successful children’s party,” he whispered. Molly took a bite. “Not a word.”
created_at
"2026-09-23T16:25:12.86Z"
9/23/2026, 4:25:12 PM
model
"openai/gpt-6-sol-20260922"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2409
moderation_latency
(null)
generation_time
13730
tokens_prompt
75
tokens_completion
841
native_tokens_prompt
66
native_tokens_completion
665
native_tokens_completion_images
0
native_tokens_reasoning
0
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
0.006782
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790180712-PEi7nQLXjRnOf9jwt49d"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790180712-B72Qh6d0XVTKWVIGM1Zz"
upstream_id
"resp_0a8d884def548a18016ab3fd68f53087d1a2ed559c38586667"
provider_responses
0
endpoint_id
"1cfc7d9d-4404-4b8e-9ee4-58ae45c9dcd4"
id
"resp_0a8d884def548a18016ab3fd68f53087d1a2ed559c38586667"
is_byok
false
latency
1219
model_permaslug
"openai/gpt-6-sol-20260922"
provider_name
"OpenAI"
status
200
total_cost
0.006782
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
68.10%
Matches word count
n/a
neededClean
false
words
486
7.48%
Dialogue to Total Word Ratio
Ratio: 42.80%, Deviation: 7.20%
neededClean
false
wordsTotal
486
wordsDialogue
208
37.7902%