NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-06-01KK1H1W0B4CBT33RY4AJYHPDR
Run Details
Model:
GPT-5.4 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
8.8s
Cost:
$0.005100
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
97.63%
Matches word count
(details)
72.29%
Dialogue to Total Word Ratio
(details)
84.9603%
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 hedge, binoculars aimed at a backyard exploding with balloons, streamers, and the shriek of sugar-powered children. Bill lay beside her, flat in the petunias, wearing a delivery cap he’d found in a puddle. “Tell me again,” Bill whispered, “why we’re robbing a seven-year-old?” “We’re not robbing a child,” Molly said. “We’re liberating high-end toys from inattentive adults.” “That sounds like robbing a child with extra syllables.” She lowered the binoculars. “Observe the patio. Mountain of presents. No guard dog. One clown, and he looks union.” “I hate clowns,” Bill muttered. “They’ve got the shoes of liars.” Molly unfolded a crayon-colored map. “You go in as the magician’s assistant.” “There is no magician.” “Exactly. Confidence is magic.” Bill squinted at the yard. “What’s my exit?” “Cake table, left gate, alley, van. I’ll create a diversion.” “With what?” She smiled. “Bouncy castle deflation.” Bill stared. “Molly, that’s not a diversion. That’s an uprising.” From the yard came a roar of children. A piñata burst like a tiny apocalypse. “Timing,” Molly said, rising. “When chaos peaks, we harvest.” Bill adjusted the soggy cap. “If I get tackled by a parent in a paper hat, avenge me.” “Professionally,” Molly said, already moving, “I’ll steal your share too.”
created_at
"2026-03-06T12:13:13.106Z"
3/6/2026, 12:13:13 PM
model
"openai/gpt-5.4-20260305"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
600
moderation_latency
(null)
generation_time
8708
tokens_prompt
77
tokens_completion
396
native_tokens_prompt
66
native_tokens_completion
329
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
0.0051
router
(null)
provider_responses
0
endpoint_id
"9ff5625c-403f-4d7f-b895-58ac7295062c"
id
"resp_0e81090047126eb00169aac4d928408196aaff84458f77bf0b"
is_byok
false
latency
138
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-1772799193-U4AMexGuhFrp0wIrnr6h"
upstream_id
"resp_0e81090047126eb00169aac4d928408196aaff84458f77bf0b"
total_cost
0.0051
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
97.63%
Matches word count
n/a
neededClean
false
words
207
72.29%
Dialogue to Total Word Ratio
Ratio: 57.55%, Deviation: 7.55%
neededClean
false
wordsTotal
212
wordsDialogue
122
84.9603%