NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM08GE8RHRTX96BZFH3V0QM9
Run Details
Model:
GPT-5.4 Mini
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
2.3s
Cost:
$0.001471
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
20.19%
Matches word count
(details)
0.61%
Dialogue to Total Word Ratio
(details)
10.4019%
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
“Bill,” Molly whispered, peering through the hedge at the brightly painted house, “tell me again why we’re infiltrating a balloon fortress full of six-year-olds?” Bill adjusted his gloves with solemnity. “Because, Molly, the evidence suggests that inside those flimsy wrapping papers lie the spoils of the season.” “You mean toys,” she said. “I mean assets.” A squeal burst from the open window, followed by the thump of tiny feet and a chorus of off-key singing. Molly flinched. “There are too many of them.” “Children are noisy, unpredictable, and small,” Bill said. “Three advantages. They won’t suspect a thing.” “One of them just licked the fence.” “Excellent surveillance technique.” Molly glared. “And what’s the plan when they ask why we’re carrying a sack the size of a body?” “We say we’re the entertainment,” Bill said, with the confidence of a man who had never entertained anyone. From inside, a voice boomed, “Party games! Everyone to the living room!” Bill straightened. “Timing is critical. We slip in with the crowd, collect the presents, and vanish before cake.” “Before cake,” Molly repeated. “That’s the line?” “That’s the line between glory and sticky disaster.” She sighed, then hefted the sack. “If this goes wrong, I’m blaming you.” Bill gave a bow. “Naturally. Now smile. We’re about to commit the most festive burglary in suburbia.”
created_at
"2026-03-18T10:40:20.776Z"
3/18/2026, 10:40:20 AM
model
"openai/gpt-5.4-mini-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
396
moderation_latency
(null)
generation_time
2098
tokens_prompt
77
tokens_completion
401
native_tokens_prompt
66
native_tokens_completion
316
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
0.0014715
router
(null)
provider_responses
0
endpoint_id
"9ee065b2-3d1c-43bc-bdd7-28af3b148282"
id
"resp_0cfedd371d30402c0169ba8114d4b081938a8eaf44d3f8ab15"
is_byok
false
latency
193
model_permaslug
"openai/gpt-5.4-mini-20260317"
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-1773830420-toH8S60CNN1ScaBoN5Q0"
upstream_id
"resp_0cfedd371d30402c0169ba8114d4b081938a8eaf44d3f8ab15"
total_cost
0.0014715
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
20.19%
Matches word count
n/a
neededClean
false
words
220
0.61%
Dialogue to Total Word Ratio
Ratio: 65.02%, Deviation: 15.02%
neededClean
false
wordsTotal
223
wordsDialogue
145
10.4019%