NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-06-01KK1GR36G5JW9MWTT7ABW49A7
Run Details
Model:
GPT-5.4
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
10.2s
Cost:
$0.004860
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
68.10%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
34.0512%
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 flattened the crayon-colored invitation on the café table and tapped the glittery clown with a gloved finger. Outside, rain lacquered the street; inside, Bill studied the room as if expecting a bouncer in a balloon hat. “Tell me again,” Bill murmured, “why we’re robbing a seven-year-old?” “Not robbing a seven-year-old,” Molly said. “Redistributing a mountain of badly wrapped consumer optimism.” Bill peered at the card. “At a trampoline park. Full of shrieking witnesses. Sticky witnesses.” “We blend in,” Molly said. “You’re an uncle. I’m a frazzled mum. We arrive late, apologize loudly, and head straight for the gift table.” “With what child?” Molly smiled. “Borrowed niece. Inflatable, if necessary.” Bill nearly dropped his teacup. “You cannot smuggle an inflatable child into a birthday party.” “Then a cardboard one. Focus. Presents get stacked by the cake, near the emergency exit. While the magician makes a rabbit suffer, you wheel the gifts out in the stroller.” “A stroller?” “Classic misdirection. No one questions a stroller.” “They question one groaning with board games.” “Then we cover them with a blanket.” Bill rubbed his temple. “And if someone asks which kid we belong to?” Molly folded the invitation and slipped it away. “We look offended and say, ‘Honestly, in this economy, shouldn’t all children belong to everyone?’”
created_at
"2026-03-06T12:07:52.791Z"
3/6/2026, 12:07:52 PM
model
"openai/gpt-5.4-20260305"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
421
moderation_latency
(null)
generation_time
10202
tokens_prompt
77
tokens_completion
394
native_tokens_prompt
66
native_tokens_completion
313
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
0.00486
router
(null)
provider_responses
0
endpoint_id
"9ff5625c-403f-4d7f-b895-58ac7295062c"
id
"resp_0ce90298910e3e200169aac398d8548194a90772a390484eef"
is_byok
false
latency
146
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-1772798872-rWGYR52eJ5YvdVqi5nNW"
upstream_id
"resp_0ce90298910e3e200169aac398d8548194a90772a390484eef"
total_cost
0.00486
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
68.10%
Matches word count
n/a
neededClean
false
words
214
0.00%
Dialogue to Total Word Ratio
Ratio: 68.95%, Deviation: 18.95%
neededClean
false
wordsTotal
219
wordsDialogue
151
34.0512%