NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM0A43DNM2FW2CMXSG2EJ94W
Run Details
Model:
GPT-5.4 Nano (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
17.1s
Cost:
$0.002366
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
60.67%
Dialogue to Total Word Ratio
(details)
80.3333%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 90% 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
"Bill, are we really doing the ‘kidnap the gifts’ routine again?" Molly whispered. "Yep," Bill said, "because adults buy boring stuff and children buy glittery chaos." "That’s not a routine," Molly argued, "that’s a lifestyle of petty retail thievery." "Language," Bill sighed. "We call it festive reallocation." "Festive reallocation sounds like a tax form," Molly said. "Hush. The party starts in ten minutes and the clown is already late." "I hate clowns," Bill admitted. "They smell like optimism and cheap eyeliner." Molly leaned closer. "Then we wear capes and become… very suspicious invisible adults." "Invisibility is expensive," Bill said, patting his empty pockets. "We’ll improvise." "With what, magic?" "With snacks," Bill said. "Kids love snacks more than morals." "That is depressingly true," Molly said. "Okay, you distract them with balloon animals." "I’m not approved for balloon duty," Bill protested. "Last time I tried, a giraffe unionized." Molly grinned. "Fine. You distract. I grab presents, and we leave before someone asks our age." "What if they ask?" Bill asked. "We’ll say we’re… ageless," Molly replied. "Like villains." "Villains with receiptless joy," Bill said. "Exactly. Now shush—someone’s coming." "And if the doorbell rings," Molly murmured, "you tell them you’re the present delivery!"
created_at
"2026-03-18T11:08:33.604Z"
3/18/2026, 11:08:33 AM
model
"openai/gpt-5.4-nano-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
16521
moderation_latency
(null)
generation_time
16810
tokens_prompt
77
tokens_completion
897
native_tokens_prompt
66
native_tokens_completion
1882
native_tokens_completion_images
0
native_tokens_reasoning
1555
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
0.0023657
router
(null)
provider_responses
0
endpoint_id
"0c835f2e-c18d-4e8c-b245-e1e3bd08b97f"
id
"resp_09820ddfb63fcf010169ba87b1af2c819495d0dff823376021"
is_byok
false
latency
250
model_permaslug
"openai/gpt-5.4-nano-20260317"
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-1773832113-kAZB3lsUoi5K4OqXgQlB"
upstream_id
"resp_09820ddfb63fcf010169ba87b1af2c819495d0dff823376021"
total_cost
0.0023657
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
60.67%
Dialogue to Total Word Ratio
Ratio: 81.59%, Deviation: 8.41%
neededClean
false
wordsTotal
201
wordsDialogue
164
80.3333%