NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-09-01KK9Q47QRFM723097MCDVC9V1
Run Details
Model:
GPT-5.4 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
22.6s
Cost:
$0.021225
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
78.49%
Dialogue to Total Word Ratio
(details)
89.2468%
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, I’m telling you, balloons are our camouflage. Nobody suspects criminals carrying balloons.” “Bill, nobody suspects clowns either, and I refuse to be arrested in shoes the size of canoes.” “Fine. Magicians?” “Too much applause. We need quiet.” “It’s a children’s party. Quiet left with the first sugar cube.” “Then we slip in as entertainment, smile at the cake, pocket the presents, and vanish before the parents notice.” “Parents always notice. They can smell mischief like bloodhounds in cardigans.” “Not if I distract them with a raffle.” “With what prize?” “Their own toaster.” “Bill, that is not a raffle. That’s confusion with tickets.” “Exactly. Meanwhile, you sweep the gift table.” “And what if the birthday child sees me?” “You tell him you’re the Fairy Godmother of Redistribution.” “He’s six, Bill, not Parliament.” Bill adjusted his cap. “All right, revised plan: I juggle, you grab, we exit through the bouncy castle.” “The bouncy castle?” “No footprints.” “Only dignity-related injuries.” “I’m willing to sacrifice dignity.” “You already did, darling, when you suggested balloons.” Molly sighed. “Right. We go in smiling, come out laden, and under no circumstances eat the jelly.” “Why?” “Last time you confessed to a piñata.” “And sang to police.”
created_at
"2026-03-09T16:33:17.57Z"
3/9/2026, 4:33:17 PM
model
"openai/gpt-5.4-20260305"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
21944
moderation_latency
(null)
generation_time
22395
tokens_prompt
77
tokens_completion
501
native_tokens_prompt
66
native_tokens_completion
1404
native_tokens_completion_images
0
native_tokens_reasoning
1107
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
0.021225
router
(null)
provider_responses
0
endpoint_id
"9ff5625c-403f-4d7f-b895-58ac7295062c"
id
"resp_058508792c1903440169aef64da81c819491cea24a74b8ab9c"
is_byok
false
latency
197
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-1773073997-vUqBb3JZs5Fs5qOHspS4"
upstream_id
"resp_058508792c1903440169aef64da81c819491cea24a74b8ab9c"
total_cost
0.021225
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
200
78.49%
Dialogue to Total Word Ratio
Ratio: 97.01%, Deviation: 7.01%
neededClean
false
wordsTotal
201
wordsDialogue
195
89.2468%