NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-10-05-01M46BKCREAQV80RQM7CWHWEHD
Run Details
Model:
GPT-6.1 Sol (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
49.4s
Cost:
$0.020702
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
100.00%
Dialogue to Total Word Ratio
(details)
100.0000%
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 the birthday invitation across their kitchen table as though it were a priceless map. Bill leaned over it, chewing a pencil. Behind them, their getaway wagon waited beside the refrigerator, its wheels squeaking with the confidence of four mice announcing a burglary to an unusually attentive neighborhood cat. “The target turns six tomorrow. Security consists of one aunt and a suspicious inflatable dinosaur.” “I refuse to fight another dinosaur. Last time the zipper caught my entirely innocent moustache.” The invitation promised magic, face painting, and unlimited squash. Bill studied this last detail gravely. Their previous caper had ended when he mistook concentrated cordial for a refreshing beverage and spent forty minutes vibrating behind a hedge. Molly had subsequently added hydration protocols to their increasingly depressing criminal operations manual. “We enter as entertainers, gather every present, then leave before anyone notices the balloon animals.” “My balloon animals are excellent. That giraffe suffered from an unconventional approach to structural integrity.” She produced two clown noses from a drawer normally reserved for overdue bills. His was red; hers sparkled with tiny silver stars. Bill regarded them with distaste. Nothing undermined professional menace like breathing through a cherry tomato, especially one that squeaked whenever its wearer attempted to discuss transport or profit. “Why must we wear noses? Surely matching waistcoats would communicate whimsy without compromising our dignity.” “Because the invitation specifically requests clowns, Bill. Your dignity was repossessed with the van Thursday.” Outside, rain tickled the window. Molly sketched a birthday cake on the back of an electricity demand, then drew arrows towards a mountain of presents. Bill added a moustache to the cake. She crossed it out. He added another, smaller moustache beneath it, like a cake maintaining two secret identities. “What if the presents contain educational toys? We cannot fence arithmetic to our usual customers.” “Then we diversify. Flashcards are compact, valuable, and unlikely to bite, unlike your ferret collection.” Bill opened their equipment bag. Inside lay confetti, a rubber chicken, three disguises, and a sandwich marked EMERGENCY. Molly inspected the chicken, squeezed it once, and winced. Its shriek suggested either imminent betrayal or a clarinet being murdered. Their occupation, she reflected, had acquired a worrying amount of poultry lately. “There's another problem, Molly. Children ask questions. Relentlessly. Last week one demanded my employment history.” “Just smile and juggle. Nobody interrogates a man juggling unless he drops something on somebody.” Molly checked the invitation again. Her expression changed. Bill recognized that expression. It usually preceded either an unexpected expense or their names appearing on laminated posters. She turned the card over. Beneath the cheerful balloons, the host had written that all gifts would be donated to the local children's hospital. For a moment, only the chicken seemed prepared to comment. Then Molly folded the plan. Bill quietly removed the emergency sandwich. They spent the evening wrapping their stolen toaster and practicing balloon dogs. By midnight, they had produced nineteen sausages, one convincing worm, and something resembling a deeply disappointed magistrate.
created_at
"2026-10-05T15:40:59.029Z"
10/5/2026, 3:40:59 PM
model
"openai/gpt-6.1-sol-20260929"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
9961
moderation_latency
(null)
generation_time
49348
tokens_prompt
75
tokens_completion
1150
native_tokens_prompt
66
native_tokens_completion
2057
native_tokens_completion_images
0
native_tokens_reasoning
1425
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
0.020702
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1791214859-MKfEFZGpoVTxlOHlRLJZ"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1791214859-ItwjywacFRiS94YgchML"
upstream_id
"resp_0174c9a557dcdf02016ac3c50b262487d1acfd682b480c5fcf"
provider_responses
0
endpoint_id
"dfb9d5c3-63f9-4263-8b78-4fa97dcb3d5e"
id
"resp_0174c9a557dcdf02016ac3c50b262487d1acfd682b480c5fcf"
is_byok
false
latency
387
model_permaslug
"openai/gpt-6.1-sol-20260929"
provider_name
"OpenAI"
status
200
total_cost
0.020702
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
500
100.00%
Dialogue to Total Word Ratio
Ratio: 30.00%, Deviation: 0.00%
neededClean
false
wordsTotal
500
wordsDialogue
150
100.0000%