NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-06-01KK1HASXE51N8R9YQEQHG070C
Run Details
Model:
GPT-5.4 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
10.0s
Cost:
$0.005040
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
0.21%
Matches word count
(details)
99.91%
Dialogue to Total Word Ratio
(details)
50.0613%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 50% 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 crouched behind the garden wall, peering at balloons bobbing above the hedge like gaudy surveillance devices. Bill unfolded a crayon-scribbled map with the gravity of an admiral plotting war, though it had clearly once belonged to a child named Poppy. “Front gate’s impossible,” Molly whispered. “Too many parents, too many casseroles.” “Then we go through the side fence,” Bill said. “I’ve bribed the dog with sausages.” “You bribed a Labrador. That’s not strategy, Bill, that’s catering.” “Mock me now, thank me when we’re knee-deep in wrapping paper.” Molly adjusted her black gloves, which made her look less like a master thief and more like a very determined mime. From the lawn came shrieks, a magician’s exhausted applause line, and the unmistakable thud of a bouncy castle losing dignity. “The presents are stacked by the cake table,” she said. “We get in during ‘Happy Birthday.’ No one hears crime over group singing.” “And if they do?” “We smile, say we’re cousins, and carry off the loot with tragic confidence.” Bill nodded. “What about the clown?” “I fear no clown.” “You should. He’s got the eyes of a tax inspector.” They rose together as the first notes of birthday song drifted over the hedge. “Ready?” Molly asked. Bill grinned. “For glory, greed, and moderately expensive toys.” “Keep low,” she said. “Never,” he replied, and immediately tripped over a gnome.
created_at
"2026-03-06T12:18:05.878Z"
3/6/2026, 12:18:05 PM
model
"openai/gpt-5.4-20260305"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
426
moderation_latency
(null)
generation_time
9914
tokens_prompt
77
tokens_completion
408
native_tokens_prompt
66
native_tokens_completion
325
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
0.00504
router
(null)
provider_responses
0
endpoint_id
"9ff5625c-403f-4d7f-b895-58ac7295062c"
id
"resp_08210ed990f4e6720169aac5fdf5a08197a3cfae502647a1ae"
is_byok
false
latency
167
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-1772799485-XNT79AmZrsnPR5SgAu9Y"
upstream_id
"resp_08210ed990f4e6720169aac5fdf5a08197a3cfae502647a1ae"
total_cost
0.00504
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
0.21%
Matches word count
n/a
neededClean
false
words
228
99.91%
Dialogue to Total Word Ratio
Ratio: 48.26%, Deviation: 1.74%
neededClean
false
wordsTotal
230
wordsDialogue
111
50.0613%