NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2025-04-16-01JRZYVVPB36NNNQWTGJC8GK5C
Run Details
Model:
o4 Mini High
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
1m 6s
Cost:
$0.035915
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
97.80%
Dialogue to Total Word Ratio
(details)
98.9002%
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
Night. Two thieves crouch outside suburban house. Molly: "The party's inside, kids squealing like a pack of sugar-high raccoons. We in?" Bill: "I brought my glitter gun. It shoots confetti. Perfect distraction." Molly: "Confetti’s cheap. We need something bigger. How about fake cake?" Bill: "Cake? It's dry. Kids don't care if it’s real. They just smash it." Molly: "Exactly. We swap the cake, wait for chaos, then slip to the gift table." Bill: "Table loaded: dolls, remote cars, that unicorn everything. Jackpot." Molly: "Time it with the magician’s rabbit reveal. Everyone gasps, we grab the haul." Bill: "What if the rabbit jumps on our heads?" Molly: "We run. Plan B: glitter bombs. Flash, bloom of sparkle, boom—confusion." Bill: "And we dive into the wrapping paper pile." Molly: "Like blind mice. Then deliver everything to our fence sale tomorrow." Bill: "Kids will cry, parents will blame magic party ghosts." Molly: "Perfect alibi. Squeaky shoes off, gloves on, window’s unlocked." Bill: "You got the crowbar? Pink sparkles? No?" Molly: "Borrowed your neon toolkit. Let's go, partner." Bill: "Remember the flank exit through the ball pit?" Molly: "Kids will dive in. We slip through the drain. Exit stage left." They slip through door.
id
"gen-1744829280-Zs2rW3jxI7XEc2DaE25E"
upstream_id
"chatcmpl-BN25gnKdX5VkSUKTF0ugS3Za83PVX"
total_cost
0.035915
cache_discount
(null)
provider_name
"OpenAI"
created_at
"2025-04-16T18:49:06.047136+00:00"
4/16/2025, 6:49:06 PM
model
"openai/o4-mini-high-2025-04-16"
app_id
182717
streamed
true
cancelled
false
latency
63177
moderation_latency
111
generation_time
65556
tokens_prompt
68
tokens_completion
8130
native_tokens_prompt
66
native_tokens_completion
8146
native_tokens_reasoning
7808
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
0.035915
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
201
97.80%
Dialogue to Total Word Ratio
Ratio: 86.14%, Deviation: 3.86%
neededClean
false
wordsTotal
202
wordsDialogue
174
98.9002%