NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2025-04-16-01JRZZF2DM2HXS75X6QRD5T65Y
Run Details
Model:
o4 Mini High
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
1m 27s
Cost:
$0.033794
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
99.74%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
49.8722%
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
Molly and Bill crouch behind a row of flamingos outside a house. Balloons bob by the window, laughter drifts out. Bill: Do we have the blueprint of the cake table? Molly: I photocopied the invitation — the seating chart doubles as a treasure map. Bill: Perfect. Which presents are we targeting first? Molly: The big blue box by the rainbow unicorn piñata. It's calling our names. Bill: What if the parents catch us? Molly: We’ll wear clown noses, join the conga line, distract them with slapstick. Bill: Then we'd look suspiciously skilled at slapstick. Molly: Exactly! Who honestly even steals with pratfalls? Bill: Smart. How do we smuggle a five-foot teddy bear? Molly: We collapse it and hide it under the giant sandwich platter. Bill: You think anyone will check under sandwiches? Molly: Nobody’s brave enough to lift a two-foot high mound of bologna. Bill: All right. At ten o’clock we slip in dressed as balloon animals. Molly: And at ten-thirty we’re gone with every gift and zero crumbs on our suits. Bill: What about the piñata? We can hit it and scoop up the loot. Molly: Only if we replace the candy with gummy frogs—they’re soft enough to hide quickly. Bill: High five.
id
"gen-1744829909-S3G4WdEiBQaRIvxOvCks"
upstream_id
"chatcmpl-BN2Fp2BK6pM4dkGRsfhfiQaLZYXs6"
total_cost
0.0337942
cache_discount
(null)
provider_name
"OpenAI"
created_at
"2025-04-16T18:59:56.831979+00:00"
4/16/2025, 6:59:56 PM
model
"openai/o4-mini-high-2025-04-16"
app_id
182717
streamed
true
cancelled
false
latency
83468
moderation_latency
127
generation_time
86390
tokens_prompt
68
tokens_completion
7647
native_tokens_prompt
66
native_tokens_completion
7664
native_tokens_reasoning
7360
num_media_prompt
(null)
num_media_completion
(null)
num_search_results
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
usage
0.0337942
Evaluation details
Result
Evaluator
Details
Meta Data
99.74%
Matches word count
n/a
neededClean
false
words
204
0.00%
Dialogue to Total Word Ratio
Ratio: 0.00%, Deviation: 90.00%
neededClean
false
wordsTotal
207
wordsDialogue
0
49.8722%