NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M37HQ20SG9C80NGGF28A73YZ
Run Details
Model:
GPT-6 Sol (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
19.6s
Cost:
$0.008962
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
43.38%
Matches word count
(details)
20.72%
Dialogue to Total Word Ratio
(details)
32.0495%
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
Molly spread a crayon drawing across the bonnet of their van. A house leaned sideways beneath a purple sun. Around it, stick figures waved what might have been swords. “Where did you get the floor plan?” Bill asked. “Birthday invitation. That’s the house on the front.” “It’s drawn by a child.” “A child who lives there. That’s an inside source.” “Those could be windows. Or teeth.” “We’ll find out. Party starts at two. At two fifteen, the magician goes on. We enter through the kitchen, take the presents, and leave through the back gate.” “Why does the magician help?” “Distraction.” “No, I mean why did you write ‘magician helps’?” Molly pressed a finger to the drawing. “I thought I’d booked him.” “You thought?” “He’s confirmed now. I booked him this morning.” “For what?” “Our getaway. He drives a van.” Bill looked at their van, which said DAD’S PLUMBING on the side in removable letters. “We also drive a van.” “He has a smoke machine.” “In a children’s party?” “He says it’s suitable for children.” “Was that before or after you told him what we’re doing?” “I haven’t told him. That’s your job.” A gust lifted the drawing. Bill pinned it with an elbow. “I’m not telling a magician we’re stealing birthday presents.” “Tell him we’re moving them.” “He’ll ask where.” “To the van.” “He’ll ask why.” “Because they’re heavy, Bill. Have you never moved anything?” “I once moved a piano.” “And did it interrogate you?” “The owner did.” Molly reached into the van and produced two matching yellow T-shirts. On each, a smiling balloon wore sunglasses above the words PARTY SUPPORT CREW. “Uniforms,” she said. “Wear yours.” “Mine says PARTY SUPP ROT CREW.” “I made a mistake with the iron.” “You also melted a hole through the balloon.” “He looks tough. Children respect that.” Bill held the shirt against his chest. “What if the parents ask who hired us?” “Say Mrs. Peabody.” “Who’s Mrs. Peabody?” “Birthday boy’s grandmother. She’s paying for the magician.” “How do you know?” “She told me when I rang her.” “You rang the grandmother?” “To confirm the magician.” “And she didn’t question that?” “She did. Quite a lot. But then she gave me the address.” Bill studied the drawing again. “Molly. This invitation says Saturday the sixteenth.” “Yes.” “Today’s the fifteenth.” Molly went still. From beyond the garden wall came shrieks, applause, and the unmistakable squeak of a balloon losing an argument. “I rang the grandmother,” she said slowly, “and told her the magician would be there today.” “So whose party is that?” They peered over the wall. Six dogs in paper crowns chased a man in a tuxedo. A smoke machine coughed beside a table heaped with wrapped gifts. Bill grinned. “New plan?” Molly watched a spaniel tear open a box and run off with a rubber duck. “Same plan,” she said. “But we’re taking the magician first.”
created_at
"2026-09-23T16:31:23.168Z"
9/23/2026, 4:31:23 PM
model
"openai/gpt-6-sol-20260922"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
6219
moderation_latency
(null)
generation_time
19573
tokens_prompt
75
tokens_completion
1087
native_tokens_prompt
66
native_tokens_completion
883
native_tokens_completion_images
0
native_tokens_reasoning
208
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
0.008962
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790181083-a48SKcMevWxZfKNNGBai"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790181083-5CYWKNyyPGn0iNWl5DRS"
upstream_id
"resp_07e520e437cce4ac016ab3fedb474487d18be56a3de12a6318"
provider_responses
0
endpoint_id
"1cfc7d9d-4404-4b8e-9ee4-58ae45c9dcd4"
id
"resp_07e520e437cce4ac016ab3fedb474487d18be56a3de12a6318"
is_byok
false
latency
1217
model_permaslug
"openai/gpt-6-sol-20260922"
provider_name
"OpenAI"
status
200
total_cost
0.008962
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
43.38%
Matches word count
n/a
neededClean
false
words
483
20.72%
Dialogue to Total Word Ratio
Ratio: 64.39%, Deviation: 5.61%
neededClean
false
wordsTotal
483
wordsDialogue
311
32.0495%