NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M37HT95DZDQM0P9V18TDVQE4
Run Details
Model:
GPT-6 Sol
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
14.6s
Cost:
$0.007062
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
11.48%
Dialogue to Total Word Ratio
(details)
5.7389%
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
Molly spread a floor plan across the café table and pinned its corners with four sugar packets. The plan showed a living room, a kitchen, and a large circle labeled CAKE. Bill leaned over it, his ski mask folded on his head like a woolly crown. “Why does the cake have its own room?” he asked. “That’s the trampoline,” Molly said. “The cake is the smaller circle.” Bill squinted. “What’s the big arrow?” “Grandma. She moves constantly.” They had robbed jewelry stores, a yacht, and, once, a vending machine that turned out to be empty. But this job had come from a professional rival who claimed nobody could steal every present from a seven-year-old’s birthday party. Molly disliked the rival and the word nobody. At three o’clock, the guests would assemble in the garden for a magician. The presents would remain in the living room, unattended for eleven minutes. Molly’s plan required a distraction, a laundry cart, and matching paper hats. Bill took a sip of coffee. “We’re stealing from children.” “We’re stealing from a pile,” Molly said. “That pile belongs to children.” “Then think of it as a very short-term relocation.” He considered this. “Will we give them back?” “Of course. Before the magician finishes.” Bill relaxed. “Right. To win the bet.” “And to avoid becoming villains in a newspaper with a photograph of your face.” Bill pulled down his ski mask. The waitress, approaching with two muffins, stopped. He pushed it up again and smiled. She set the muffins on another table. Molly outlined the route with a pencil. They would enter through the side gate, cross behind the inflatable castle, and reach the living room through a sliding door. Bill would load the gifts while Molly occupied Grandma, whose movements suggested she had been assigned security. At two fifty-eight, they found the garden gate guarded by a boy in a cardboard knight’s helmet. His sword was a pool noodle. Molly showed him the paper hats. “Party crew,” she said. The boy pointed his noodle at Bill. “What’s your magic trick?” Bill looked at Molly. She looked at the laundry cart. “He makes things disappear,” Molly said. “Show me.” Bill produced a coin, fumbled it, and watched it roll under the gate. The boy opened the gate to retrieve it. Molly wheeled the cart through. In the garden, the magician announced a vanishing act. Every adult turned toward him except Grandma. She stood at the sliding door, holding a plate of sandwiches. “You’re late,” she said. Molly froze. Bill froze harder. Grandma handed them the plate. “Put these by the drinks, then take the presents inside. We’re doing the opening after cake.” Molly and Bill carried the gifts from the living room into the laundry cart under Grandma’s brisk supervision. The pile barely fit. When the magician produced a rabbit, the boy in the helmet spotted them wheeling it across the patio. “They made the presents disappear!” he shouted. Grandma checked the cart, counted the gifts, and nodded. “They’re in the kitchen, dear. Now wash your hands.” Molly glanced at Bill. “Eleven minutes?” “Seven,” he whispered. She grinned. They had won the bet. Grandma made them serve cake for the other four.
created_at
"2026-09-23T16:33:08.79Z"
9/23/2026, 4:33:08 PM
model
"openai/gpt-6-sol-20260922"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2838
moderation_latency
(null)
generation_time
14523
tokens_prompt
75
tokens_completion
858
native_tokens_prompt
66
native_tokens_completion
693
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
0.007062
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790181188-Io8oCrVXUjWS5JT9GIGr"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790181188-5GXepE8SzX3ybd2fL41S"
upstream_id
"resp_0a113448f54801ef016ab3ff44eaec87d1a2ef1ff1b91887e0"
provider_responses
0
endpoint_id
"1cfc7d9d-4404-4b8e-9ee4-58ae45c9dcd4"
id
"resp_0a113448f54801ef016ab3ff44eaec87d1a2ef1ff1b91887e0"
is_byok
false
latency
1143
model_permaslug
"openai/gpt-6-sol-20260922"
provider_name
"OpenAI"
status
200
total_cost
0.007062
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
0.00%
Matches word count
n/a
neededClean
false
words
534
11.48%
Dialogue to Total Word Ratio
Ratio: 23.42%, Deviation: 6.58%
neededClean
false
wordsTotal
538
wordsDialogue
126
5.7389%