NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-06-01KK1J7198S9YXAAT1DMJWJV5Y
Run Details
Model:
GPT-5.4 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
11.1s
Cost:
$0.004830
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
68.10%
Matches word count
(details)
69.09%
Dialogue to Total Word Ratio
(details)
68.5947%
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 blueprint of the bungalow on a sticky café table, pinning one corner with a spoon and the other with Bill’s trembling hand. Outside, balloons bobbed over the garden wall like cheerful surveillance devices. “Children’s parties are chaos with frosting,” Molly said. “Perfect cover.” Bill peered at the sketch. “I still say we’ll be undone by a clown.” “The clown is my alibi.” “Your alibi rides a unicycle?” “And makes balloon poodles. Focus. Presents stacked by the fireplace, cake at two, screaming at two-oh-one.” Bill adjusted his cap. “What about the parents?” “Distracted by dietary requirements and passive-aggressive praise.” He nodded gravely. “Savage.” From the wall came a burst of tinny music and the shriek of delight that always sounded, to Bill, a little like an air-raid siren. “We go in with the magician,” Molly said. “You carry the prop trunk.” “I hate prop trunks. They always smell of doves and disappointment.” “Then think of loot. Board games, drones, that suspiciously expensive dollhouse.” Bill’s eyes brightened. “Right. In, swipe, out.” “No greed,” Molly warned. “If it squeaks, sings, or requires batteries the size of soup cans, leave it.” “And if a child spots us?” Molly folded the blueprint with priestly calm. “Smile kindly,” she said. “Then ask where the good presents are.”
created_at
"2026-03-06T12:33:30.929Z"
3/6/2026, 12:33:30 PM
model
"openai/gpt-5.4-20260305"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
486
moderation_latency
(null)
generation_time
11067
tokens_prompt
77
tokens_completion
392
native_tokens_prompt
66
native_tokens_completion
311
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
0.00483
router
(null)
provider_responses
0
endpoint_id
"9ff5625c-403f-4d7f-b895-58ac7295062c"
id
"resp_055af822660e22810169aac99b01708193bfe50fa67ad7f125"
is_byok
false
latency
151
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-1772800410-5XqUbe256yiUSeF1x5E4"
upstream_id
"resp_055af822660e22810169aac99b01708193bfe50fa67ad7f125"
total_cost
0.00483
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
69.09%
Dialogue to Total Word Ratio
Ratio: 57.80%, Deviation: 7.80%
neededClean
false
wordsTotal
218
wordsDialogue
126
68.5947%