NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-06-01KK1HA6GQMK8QHR893E1AQG7X
Run Details
Model:
GPT-5.4
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
9.8s
Cost:
$0.004515
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
27.17%
Matches word count
(details)
98.21%
Dialogue to Total Word Ratio
(details)
62.6887%
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 and Bill crouched behind a hedge shaped like a startled rabbit, studying the back garden where balloons bobbed like surveillance devices and a clown wrestled a folding table into submission. The children’s party was in full swing: shrieking, cake-smearing, a kind of cheerful anarchy that made serious crime feel almost respectable. Bill adjusted his flat cap and peered at the mountain of gifts stacked beneath a banner reading HAPPY EIGHTH BIRTHDAY, OLIVER!!! in letters large enough to shame a satellite. “Too many kids,” Bill muttered. “Too much wrapping paper,” Molly replied. She unfolded the hand-drawn plan, which looked less like a blueprint and more like a treasure map produced during a mild earthquake. Step one involved blending in with the entertainers. Step two relied heavily on confusion, a resource the average child produced in industrial quantities. Near the patio, a magician pulled a coin from his own ear and looked startled to find it there. Parents laughed nervously, already defeated. Molly smiled with professional admiration. Amateur distraction, but spirited. Bill eyed a tray of sausage rolls passing by in the hands of a harried aunt. “We go in as catering?” Molly shook her head. “No. Clowns.” A scream rose from the bouncy castle, followed by applause. Bill sighed, resigned to greasepaint, tiny hats, and larceny among juice boxes.
created_at
"2026-03-06T12:17:46.013Z"
3/6/2026, 12:17:46 PM
model
"openai/gpt-5.4-20260305"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
425
moderation_latency
(null)
generation_time
9752
tokens_prompt
77
tokens_completion
364
native_tokens_prompt
66
native_tokens_completion
290
native_tokens_completion_images
0
native_tokens_reasoning
0
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
0.004515
router
(null)
provider_responses
0
endpoint_id
"9ff5625c-403f-4d7f-b895-58ac7295062c"
id
"resp_01cf347efdfdf8600169aac5ea124081969720d59f0ba97174"
is_byok
false
latency
170
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-1772799466-lsJlPUGS8EN9wZFIONmG"
upstream_id
"resp_01cf347efdfdf8600169aac5ea124081969720d59f0ba97174"
total_cost
0.004515
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
27.17%
Matches word count
n/a
neededClean
false
words
219
98.21%
Dialogue to Total Word Ratio
Ratio: 6.33%, Deviation: 3.67%
neededClean
false
wordsTotal
221
wordsDialogue
14
62.6887%