NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-06-01KK1J44ZAV6635XVSF4H2TDZ1
Run Details
Model:
GPT-5.4 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
11.1s
Cost:
$0.005040
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
9.61%
Matches word count
(details)
0.04%
Dialogue to Total Word Ratio
(details)
4.8263%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 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
Molly flattened the crayon-splattered invitation on the café table and tapped the clown sticker like it owed her money. Outside, rain made the street shine; inside, Bill wore a balloon-animal hat he’d “borrowed” for research and looked tragically confident. “Children’s parties are chaos,” Molly said. “Chaos is camouflage.” Bill squinted. “Camouflage usually doesn’t involve a magician named Professor Sprinkles.” “It does tonight. Back gate, six-thirty. Face painter by the shed, bouncy castle to the left, gift mountain by the cake.” “Gift mountain,” Bill repeated reverently. “A summit of poorly hidden treasure.” Molly slid him a rough sketch. “Parents cluster near coffee. Kids swarm sugar. We drift in as hired entertainment.” Bill straightened the balloon hat. “I can juggle three oranges and a grudge.” “Perfect. I’ll do balloons.” “You can’t make balloons.” “I can make panic, which is basically the same thing.” He grinned. “What about the dog? Every suburban fortress has one emotionally unstable dog.” “Bribed with sausages.” “What about the clown?” “More frightened of me than I am of him.” Bill leaned closer. “And if someone asks whose children we are?” Molly folded the invitation and stood. “We look offended, say ‘the birthday one,’ and keep walking.” Bill rose too, inspired. “Molly, this is either our lowest moment or our finest work.” “Bill,” she said, “those are usually the same thing.”
created_at
"2026-03-06T12:31:56.403Z"
3/6/2026, 12:31:56 PM
model
"openai/gpt-5.4-20260305"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
416
moderation_latency
(null)
generation_time
11013
tokens_prompt
77
tokens_completion
420
native_tokens_prompt
66
native_tokens_completion
325
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)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"completed"
usage
0.00504
router
(null)
provider_responses
0
endpoint_id
"9ff5625c-403f-4d7f-b895-58ac7295062c"
id
"resp_046d4e383d97f5fb0169aac93c7ef481969ab13c6c6923aea5"
is_byok
false
latency
167
model_permaslug
"openai/gpt-5.4-20260305"
provider_name
"OpenAI"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.0; linux; x64))"
http_referer
(null)
api_type
"completions"
id
"gen-1772800316-McUorN8Vx7JeL19vTPpj"
upstream_id
"resp_046d4e383d97f5fb0169aac93c7ef481969ab13c6c6923aea5"
total_cost
0.00504
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
9.61%
Matches word count
n/a
neededClean
false
words
222
0.04%
Dialogue to Total Word Ratio
Ratio: 66.67%, Deviation: 16.67%
neededClean
false
wordsTotal
225
wordsDialogue
150
4.8263%