NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M37HRQPA6Z5P16DYTQ4MD1AJ
Run Details
Model:
GPT-6 Sol (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
22.6s
Cost:
$0.010792
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
86.38%
Matches word count
(details)
8.95%
Dialogue to Total Word Ratio
(details)
47.6668%
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 the floor plan across the café table, pinning one corner with a sugar bowl and another with Bill’s elbow. Outside, rain tapped the window. Inside, Bill studied the drawing upside down. “The presents will be stacked beside the cake table,” Molly said. “We go in, collect them, and leave before anyone notices.” Bill turned the paper around. Molly had drawn balloons, a pirate ship, and a large red X. The building itself appeared to have been an afterthought, squeezed into the remaining corner. “Is the X where the presents are?” he asked. “Or where you expect me to fall over?” “The presents. I’ve marked the cake separately. That’s the circle with the candles. Try to keep your priorities straight.” Bill looked offended. His priorities were perfectly straight: steal the presents, avoid arrest, eat cake. He had even written them on his hand that morning, though the rain had reduced “arrest” to “rest.” “What’s our cover?” he asked. “I can’t pass for somebody’s uncle. I’ve got a face parents warn children about.” “You’re the magician. I’m your assistant. We borrow the costumes, perform a trick, then make the presents disappear.” Molly produced a black cape from her bag. Its hem was dotted with silver stars; a label inside read PROPERTY OF THE LITTLE ACORNS PLAYGROUP. Bill held it up against his chest. It reached his knees. “I’m not wearing a stolen magician’s cape to steal birthday presents,” he said. “That’s too much crime in one outfit.” “You’re right. We’ll return the cape when we leave. Happy? Now, can you pull a rabbit out of a hat?” An hour later, Bill stood in the community hall beneath a banner reading HAPPY SEVENTH BIRTHDAY, PIP. He had no hat and no rabbit. Molly, wearing sequins, was carrying both, plus a surprising amount of confidence. “Ladies and gentlemen,” Bill announced to the children, “prepare to witness an act of astonishing magic. Please keep your eyes on my assistant.” “That’s not how the trick goes,” Molly whispered. “They’re supposed to look at you while I take the presents.” A girl in a paper crown stepped forward. Bill recognized her from the photograph Molly had shown him: Pip, the birthday girl. She looked from his cape to the rabbit in Molly’s arms, then held out her hands. “Are you going to give me the rabbit?” Pip asked. “Mum said the magician might have one.” “No,” Bill said. “The rabbit is working. Would you like to help me with a different trick?” Pip nodded and climbed onto the little stage. Molly edged toward the gift table, where twenty brightly wrapped boxes waited. She lifted the smallest one. A chorus of children shouted, “That’s mine!” She put it down again. “They’re all yours,” Pip explained. “It’s a swapping party. Everyone brought an old toy, and everyone gets to pick one.” Bill looked at Molly. Molly looked at the boxes. Pip picked up the smallest and handed it to Bill. Inside was a plastic wand with one bent tip. “Good,” Pip said. “Now you can do magic properly.”
created_at
"2026-09-23T16:32:18.139Z"
9/23/2026, 4:32:18 PM
model
"openai/gpt-6-sol-20260922"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
7172
moderation_latency
(null)
generation_time
22580
tokens_prompt
75
tokens_completion
1051
native_tokens_prompt
66
native_tokens_completion
1066
native_tokens_completion_images
0
native_tokens_reasoning
395
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
0.010792
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790181138-69wQVKT91gz2Smx1ypgV"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790181138-3KcrMSmjEFC5S2i7FBpf"
upstream_id
"resp_08dbb8186798ed5e016ab3ff12516887d193149cab6a982cbb"
provider_responses
0
endpoint_id
"1cfc7d9d-4404-4b8e-9ee4-58ae45c9dcd4"
id
"resp_08dbb8186798ed5e016ab3ff12516887d193149cab6a982cbb"
is_byok
false
latency
1118
model_permaslug
"openai/gpt-6-sol-20260922"
provider_name
"OpenAI"
status
200
total_cost
0.010792
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
86.38%
Matches word count
n/a
neededClean
false
words
511
8.95%
Dialogue to Total Word Ratio
Ratio: 43.05%, Deviation: 6.95%
neededClean
false
wordsTotal
511
wordsDialogue
220
47.6668%