NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M37HTXFXSRGS1PXXMYWAJDFN
Run Details
Model:
GPT-6 Sol (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
17.5s
Cost:
$0.007592
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
98.71%
Matches word count
(details)
13.20%
Dialogue to Total Word Ratio
(details)
55.9572%
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
From the hedge opposite number forty-two, Molly studied the party through a pair of opera glasses. Twelve children occupied the garden. A magician was pulling damp scarves from a hat. Beside the kitchen door stood a mountain of wrapped presents large enough to alter the local weather. “Security?” Bill asked. “One grandmother, two parents, and a magician with terrible sleeves.” “Those sleeves could hide anything.” “They’re hiding six scarves and, I suspect, his lunch.” Bill lowered himself behind the hedge. His black burglar’s outfit had been assembled from three different uniforms, leaving him looking like a waiter who had lost an argument with a chimney. Molly wore a paper crown and carried a gift bag full of empty gift bags. She considered her disguise flawless. “The plan remains simple,” she said. “We go in, join the party, and carry out the presents.” “All of them? There’s a bicycle.” “We’ll make two trips.” “I brought one bicycle helmet.” “Bill, nobody needs a helmet to carry a bicycle.” “You said there might be party games.” A burst of applause came from the garden. The magician bowed, revealing an enormous stuffed rabbit protruding from his waistband. Molly checked her watch. In seven minutes the cake would arrive, according to the invitation she had found pinned to the supermarket noticeboard. During the singing, every eye would be on the candles. Every eye except the grandmother’s, probably. Grandmothers had peripheral vision developed over decades of locating biscuits. Molly slipped through the gate first. Bill followed with the uncertain gait of a man approaching either a crime scene or a bouncy castle. Nobody challenged them. A child handed Bill a balloon, then ran off before he could explain that he preferred not to be armed. At the present table, Molly opened her largest bag. Inside, smaller bags unfurled like a magician’s trick that required considerably less practice. Bill lifted a rectangular parcel and turned it over. “Read the label,” Molly whispered. “We don’t want to leave evidence.” “It says, ‘To Oliver, from Aunt Molly.’” Molly stared at it. She looked across the garden at the birthday boy, who was wearing a cardboard crown over the same untidy curls she saw in her brother’s family photographs. She had promised to attend this party. That was why the invitation had caught her eye. “You’re Aunt Molly?” Bill said. “I may have overlooked a detail.” “Did you forget your nephew’s birthday?” “Last year. This year I brought a present.” “Where is it?” Molly glanced into her empty gift bag. The magician began a drumroll on a plastic table, and the children surged toward the cake. “Still at the shop,” she said. Bill set down Oliver’s parcel. After a moment, he took off his bicycle helmet. By the time the singing ended, he had filled it with balloons and tied on a bow. It was, Molly thought, the least practical present at the party. It was also the only one they could honestly claim.
created_at
"2026-09-23T16:33:29.605Z"
9/23/2026, 4:33:29 PM
model
"openai/gpt-6-sol-20260922"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
4992
moderation_latency
(null)
generation_time
17453
tokens_prompt
75
tokens_completion
915
native_tokens_prompt
66
native_tokens_completion
746
native_tokens_completion_images
0
native_tokens_reasoning
125
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
0.007592
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790181209-sjSbysG8CEYzS7XhZ1LK"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790181209-4h2PYRScUuUFMPGUGh9c"
upstream_id
"resp_0fa3df2820404850016ab3ff59bd9087d1ba2e7dd678222042"
provider_responses
0
endpoint_id
"1cfc7d9d-4404-4b8e-9ee4-58ae45c9dcd4"
id
"resp_0fa3df2820404850016ab3ff59bd9087d1ba2e7dd678222042"
is_byok
false
latency
1252
model_permaslug
"openai/gpt-6-sol-20260922"
provider_name
"OpenAI"
status
200
total_cost
0.007592
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
98.71%
Matches word count
n/a
neededClean
false
words
494
13.20%
Dialogue to Total Word Ratio
Ratio: 23.64%, Deviation: 6.36%
neededClean
false
wordsTotal
495
wordsDialogue
117
55.9572%