NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-09-01KK9PV5N5JGAGKVXE16759BJK
Run Details
Model:
GPT-5.4 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
40.5s
Cost:
$0.025140
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
100.00%
Dialogue to Total Word Ratio
(details)
99.9995%
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
Molly crouched behind the community hall's hedge, notebook balanced on one knee, studying the battlefield of balloons, bunting, and sugar-fueled chaos. Inside, a clown on stilts wobbled past the windows like a very cheerful emergency. Bill lay beside her in a cardigan stolen from a golf club, looking less like a master criminal than a substitute geography teacher who'd misplaced Belgium. The plan, Molly insisted, depended on timing: arrive during the cake song, slip through the kitchen, load the presents into the laundry trolley, and vanish before any parent noticed the suspicious decline in wrapping paper. It was, she admitted, the classiest robbery either of them had attempted involving a bouncy castle. “Molly, this feels morally sticky,” Bill whispered. “Bill, it's reconnaissance with ribbons,” Molly said. He watched a magician pull a dove from a hat and nearly swallow his own mustache from nerves. Somewhere, a child screamed with the ecstatic ferocity only jelly can inspire. Molly clicked her pencil, eyes shining. On the patio sat the loot table, glittering beneath fairy lights like a department store had exploded. Bill sighed, adjusted his cardigan, and prepared to burglarize joy. Bill gulped. “If toddlers tackle me, I'm surrendering and joining the cake.”
created_at
"2026-03-09T16:28:20.533Z"
3/9/2026, 4:28:20 PM
model
"openai/gpt-5.4-20260305"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
37395
moderation_latency
(null)
generation_time
40298
tokens_prompt
77
tokens_completion
731
native_tokens_prompt
66
native_tokens_completion
1665
native_tokens_completion_images
0
native_tokens_reasoning
1398
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
0.02514
router
(null)
provider_responses
0
endpoint_id
"9ff5625c-403f-4d7f-b895-58ac7295062c"
id
"resp_0252d8bca688f5290169aef524d8808194980205a4a03bf746"
is_byok
false
latency
286
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-1773073700-Nmc6FNHkS6jNDQbSxUPq"
upstream_id
"resp_0252d8bca688f5290169aef524d8808194980205a4a03bf746"
total_cost
0.02514
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
201
100.00%
Dialogue to Total Word Ratio
Ratio: 9.90%, Deviation: 0.10%
neededClean
false
wordsTotal
202
wordsDialogue
20
99.9995%