NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M36RC56HPE62348K15345CRP
Run Details
Model:
GPT-6 Luna (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
38.8s
Cost:
$0.001907
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
99.65%
Dialogue to Total Word Ratio
(details)
99.8237%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 10% 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
At sunset, Molly and Bill crouched behind a hedge outside the community hall, dressed as nervous clowns. Through the window, children swarmed around a mountain of birthday presents, while a papier-mâché dinosaur guarded the entrance with menace. Bill unfolded their floor plan, which was actually a napkin bearing a frosting stain and the words “DO NOT EAT.” “Molly, the piñata is armed,” Bill whispered. “Then distract it with your résumé,” Molly said. “Our getaway cake has wheels,” Bill said. “Excellent. Aim for frosting,” Molly replied. Molly inspected the party through binoculars, focusing on a toddler, who waved and offered her a damp cracker. She accepted it, because undercover work demanded sacrifices, and because the toddler had begun negotiations with confidence. The plan was simple: enter during “Present Parade,” substitute identical boxes, then roll the loot out beneath a blanket. Unfortunately, Bill’s shoes squeaked with each step, and one shoe had a bell that rang whenever he felt anxious. Molly tied the napkin map around his ankle, disguising the bell as a fashionable crime accessory. The dinosaur’s eyes followed them. A volunteer in a cupcake hat approached, carrying a clipboard and the authority of someone who knew where candles were kept.
created_at
"2026-09-23T09:08:31.581Z"
9/23/2026, 9:08:31 AM
model
"openai/gpt-6-luna-20260922"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
15908
moderation_latency
(null)
generation_time
38762
tokens_prompt
75
tokens_completion
1111
native_tokens_prompt
66
native_tokens_completion
3801
native_tokens_completion_images
0
native_tokens_reasoning
3537
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
0.0019071
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1790154511-gF71xtjxWlySmBgnz2l7"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790154511-76KlYXL8HOIBoI66QmXL"
upstream_id
"resp_0620b45c68b3097b016ab3970fb3b487d1ac71b51a4dec9437"
provider_responses
0
endpoint_id
"05e94e02-b9c2-4bbb-ba55-4082ee9ad687"
id
"resp_0620b45c68b3097b016ab3970fb3b487d1ac71b51a4dec9437"
is_byok
false
latency
405
model_permaslug
"openai/gpt-6-luna-20260922"
provider_name
"OpenAI"
status
200
total_cost
0.0019071
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
100.00%
Matches word count
n/a
neededClean
false
words
200
99.65%
Dialogue to Total Word Ratio
Ratio: 12.44%, Deviation: 2.44%
neededClean
false
wordsTotal
201
wordsDialogue
25
99.8237%