NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M37HQSTD3PYENS4WNAEASGN7
Run Details
Model:
GPT-6 Sol (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
16.7s
Cost:
$0.007252
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
99.38%
Matches word count
(details)
7.15%
Dialogue to Total Word Ratio
(details)
53.2637%
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
The invitation specified two o’clock, fancy dress optional, and absolutely no peanuts. Molly considered the third instruction the most intimidating. Bill had already eaten half a bag of them in the van. “Your costume looks like a tax inspector,” Molly said. “I’m the party magician.” “You’re wearing a suit and carrying a briefcase.” “The magic is that I can make deductions disappear.” Molly adjusted the paper crown perched on her head. She was supposed to be a princess, though the crown came from a takeaway and read THANK YOU FOR YOUR ORDER. Through the community hall window she could see a mountain of presents beneath a table, each wrapped in bright paper. There were enough to fill the van twice over, provided Bill moved the spare tyre, his camping chair, and the ornamental goose he insisted was lucky. “Remember the plan,” she said. “We mingle, wait for the cake, then take the presents.” “All of them?” “That is what ‘all’ means.” “Even the tiny flat one?” “Especially that one. It could be cash.” Bill peered through the window. A dozen children in capes thundered around a woman holding a tray of juice. One child had painted whiskers on his face and was chasing another with a balloon sword. Bill swallowed the last peanut and looked at Molly. “Why don’t we steal something quieter? A museum, perhaps.” “We’ve come this far.” They entered behind a family carrying a gift bag. Nobody questioned the crown or Bill’s briefcase. A small girl immediately asked him to do a trick. Bill opened the case. Inside were a sandwich, three receipts, and the ornamental goose’s missing beak. He shut it again and announced that the trick required absolute silence. The girl stared at him until he pulled a coin from behind his own ear. Molly had never seen him do that before. Judging by his expression, neither had Bill. Soon six children wanted coins. Bill found one more in his pocket, then ran out. He turned a receipt into a paper aeroplane; it flew into the cake. Molly tried to retrieve it and was handed a plastic knife, a stack of plates, and responsibility for cutting twenty-four equal slices. “This is going badly,” Bill whispered. “Keep them busy. I’ll move the presents.” “How?” “I’ll tell the adults I’m taking them to the gift table.” “That is the gift table.” “I’ll think of another table.” Before Molly could, the birthday girl climbed onto a chair and announced it was present-opening time. Every child rushed toward the pile. Molly and Bill were swept along and put to work passing parcels over their heads. They stayed through the final gift, a pair of roller skates, and helped gather the wrapping paper afterward. Outside, Molly opened the van’s rear doors. Bill loaded in the rubbish bag they had been asked to take to the bins. “We stole nothing,” he said. Molly held up a slice of cake wrapped in a napkin. “Speak for yourself.”
created_at
"2026-09-23T16:31:47.54Z"
9/23/2026, 4:31:47 PM
model
"openai/gpt-6-sol-20260922"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
3887
moderation_latency
(null)
generation_time
16641
tokens_prompt
75
tokens_completion
905
native_tokens_prompt
66
native_tokens_completion
712
native_tokens_completion_images
0
native_tokens_reasoning
94
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
0.007252
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790181107-6NXH4EyGZx6iGIbdiF9j"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790181107-SlACBBb0rGH6WjKKA5tK"
upstream_id
"resp_055537a8171e9968016ab3fef3a4c887d1902a625859a0881c"
provider_responses
0
endpoint_id
"1cfc7d9d-4404-4b8e-9ee4-58ae45c9dcd4"
id
"resp_055537a8171e9968016ab3fef3a4c887d1902a625859a0881c"
is_byok
false
latency
401
model_permaslug
"openai/gpt-6-sol-20260922"
provider_name
"OpenAI"
status
200
total_cost
0.007252
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
99.38%
Matches word count
n/a
neededClean
false
words
495
7.15%
Dialogue to Total Word Ratio
Ratio: 22.74%, Deviation: 7.26%
neededClean
false
wordsTotal
497
wordsDialogue
113
53.2637%