NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-06-01KK1HEQSYG04DVM132FQEBNAT
Run Details
Model:
GPT-5.4 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
9.3s
Cost:
$0.005025
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
2.01%
Matches word count
(details)
80.46%
Dialogue to Total Word Ratio
(details)
41.2369%
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 over the fence like colorful witnesses. Bill unfolded a crayon-smeared map stolen from a refrigerator and frowned with professional disgust. “Front gate’s suicide,” Molly whispered. “Too many parents with folding chairs and opinions.” “Back gate, then,” Bill said. “I clocked one clown, two grandmothers, and a magician who looks medically tired.” “Magicians are always dangerous.” “So are grandmothers.” From the yard came a shriek of sugar-fueled joy, followed by the thud of a small body hitting a bouncy castle. Molly adjusted her black gloves, which made her look less like a criminal mastermind and more like a glamorous dishwasher. “When the cake arrives, everyone sings,” she said. “That’s our window.” “And the presents?” “Stacked by the patio doors under the banner with the misspelled pirate.” Bill nodded solemnly. “A tragic security failure.” A terrier in a party hat trotted past them carrying a sausage roll. Both thieves froze until it disappeared round the bins. “If we’re caught,” Bill murmured, “we say we’re exotic uncles.” “You can’t be an exotic uncle from Croydon.” “I’ll do an accent.” “You’ll do prison.” Inside, children howled the opening bars of Happy Birthday. Molly grinned. “Masks on.” Bill pulled a unicorn napkin over his face. “For the honour of organized crime,” he said. “And for the Lego set,” Molly replied.
created_at
"2026-03-06T12:20:14.789Z"
3/6/2026, 12:20:14 PM
model
"openai/gpt-5.4-20260305"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
396
moderation_latency
(null)
generation_time
9264
tokens_prompt
77
tokens_completion
419
native_tokens_prompt
66
native_tokens_completion
324
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
0.005025
router
(null)
provider_responses
0
endpoint_id
"9ff5625c-403f-4d7f-b895-58ac7295062c"
id
"resp_008e3f5599b02e4c0169aac67ed6c08197b18c2e4e9c8a1803"
is_byok
false
latency
136
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-1772799614-tibh3V5Suk2VMaEs3Cf0"
upstream_id
"resp_008e3f5599b02e4c0169aac67ed6c08197b18c2e4e9c8a1803"
total_cost
0.005025
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
2.01%
Matches word count
n/a
neededClean
false
words
225
80.46%
Dialogue to Total Word Ratio
Ratio: 43.17%, Deviation: 6.83%
neededClean
false
wordsTotal
227
wordsDialogue
98
41.2369%