NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M37HV71YG2Q3WV2EYWXNX18T
Run Details
Model:
GPT-6 Sol (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
19.0s
Cost:
$0.008532
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
99.38%
Matches word count
(details)
44.64%
Dialogue to Total Word Ratio
(details)
72.0107%
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
From the bakery opposite the community hall, Molly surveyed the birthday party through binoculars shaped like opera glasses. She had bought them at a charity shop and refused to admit they were binoculars. “That’s thirty-one presents,” she said. “Thirty-two if the enormous bow in the corner has something under it. Bill, this is our greatest opportunity since the unattended cheese hamper.” Bill lowered his newspaper. He had cut two eyeholes in it, then spent twenty minutes reading a menu printed on the back. “The hamper was attended,” he said. “By a goose. I still have the scar, and you still owe me for the trousers.” Inside the hall, children swarmed around a magician whose hat appeared to contain an inexhaustible supply of scarves. The present table stood behind him, bright as a treasure chest. “We go in as party entertainers,” Molly said. “You’ll be the magician’s assistant. I’ll be the mysterious woman who carries things away while everyone applauds.” “Why do I have to be the assistant? I can do magic. Watch.” Bill produced a coin from behind Molly’s ear. It was warm and slightly sticky. “That was in your hand the entire time. Also, it’s my coin. You borrowed it for the bus and we walked here.” Bill returned the coin with a wounded expression. Molly studied the hall again. A small boy in a paper crown pressed his nose against the window and waved at her. “New complication,” she said. “The birthday monarch has spotted us. Smile like an entertainer, not like a man identifying valuables.” Bill smiled. The boy recoiled and ran off. Molly folded the opera glasses and put them away. Across the street, the magician bowed; the scarves tangled around his ankles. “We could just walk in,” Bill said. “There’s a sign on the door. It says EVERYONE WELCOME. That’s practically an invitation to criminals.” “It’s an invitation to the party. We still need a reason to leave with thirty-two parcels and not be followed by thirty-two angry children.” The hall door opened. A woman carried out a cardboard box labelled CAKE FOR THE SHELTER. Behind her, the crowned boy struggled under a sack nearly his size. Molly rose from the bakery table. “Change of plan. The presents are for the shelter, aren’t they? The children are collecting them.” “Excellent,” Bill said. “Then we’ll have an even larger haul when we intercept the shelter. Do shelters have geese?” Molly stared at him until his smile retreated. She crossed the street, Bill trailing behind, and held the door for the boy. He thanked her and pointed to a pile of empty boxes. “Can you help?” he asked. “We need to carry everything to the van.” Molly looked at Bill. His shoulders sagged beneath the weight of an opportunity he was finally clever enough to refuse. “Of course,” Molly told the boy. “My assistant loves carrying things.” Bill picked up two boxes. “For the record,” he muttered, “I was going to offer.”
created_at
"2026-09-23T16:33:39.396Z"
9/23/2026, 4:33:39 PM
model
"openai/gpt-6-sol-20260922"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
5669
moderation_latency
(null)
generation_time
19004
tokens_prompt
75
tokens_completion
917
native_tokens_prompt
66
native_tokens_completion
840
native_tokens_completion_images
0
native_tokens_reasoning
204
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
0.008532
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790181219-YhOXSgnwK96KRo7b2qXx"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790181219-rH1YWH8UOWOoUdKwitAq"
upstream_id
"resp_052c34edb5cd26a8016ab3ff637d6887d1acea68a7d22e1589"
provider_responses
0
endpoint_id
"1cfc7d9d-4404-4b8e-9ee4-58ae45c9dcd4"
id
"resp_052c34edb5cd26a8016ab3ff637d6887d1acea68a7d22e1589"
is_byok
false
latency
582
model_permaslug
"openai/gpt-6-sol-20260922"
provider_name
"OpenAI"
status
200
total_cost
0.008532
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
44.64%
Dialogue to Total Word Ratio
Ratio: 45.98%, Deviation: 4.02%
neededClean
false
wordsTotal
498
wordsDialogue
229
72.0107%