NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-09-01KK9QBJZT5111SYCS4N0A2J49
Run Details
Model:
GPT-5.4 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
37.6s
Cost:
$0.027540
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
99.74%
Matches word count
(details)
99.95%
Dialogue to Total Word Ratio
(details)
99.8458%
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 studied the bungalow through hedge gaps, where balloons bobbed like nervous sentries and a magician unloaded rabbits with union-level resentment. Bill crouched beside her, wearing a caterer’s hat so tall it made stealth theoretical. From the garden came shrieks of sugar-powered delight, the unmistakable soundtrack of loot. A pirate piñata swung over a trestle table groaning beneath wrapped presents in glossy towers. Every gift tag fluttered like a surrender flag. Molly clicked her notebook, sketched routes between cupcakes and parents, and sighed the sigh of facing glitter. Bill adjusted his moustache, sliding south with sweat. “Children guard treasure badly,” Molly whispered. “We’re not robbing children; we’re liberating surprises from poor circulation,” Bill said. “Same sin, better slogan. What’s the entry?” “Cake delivery. Everyone trusts a man carrying sponge.” “And the exit?” “Pram brigade at three o’clock. Follow the grandparents; they’re invisible.” “What about the dog?” “I’ve got sausages.” “What about the clown?” “I’ve got standards.” “We take the presents, leave the pony, and never mention glitter again.” “Deal. If caught, we cry and say we’re method actors researching disappointment.” “Fine, but if there’s a bouncy castle, I’m calling it reconnaissance and taking one dignified bounce.”
created_at
"2026-03-09T16:37:18.47Z"
3/9/2026, 4:37:18 PM
model
"openai/gpt-5.4-20260305"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
35110
moderation_latency
(null)
generation_time
37386
tokens_prompt
77
tokens_completion
370
native_tokens_prompt
66
native_tokens_completion
1825
native_tokens_completion_images
0
native_tokens_reasoning
1543
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
0.02754
router
(null)
provider_responses
0
endpoint_id
"9ff5625c-403f-4d7f-b895-58ac7295062c"
id
"resp_0eb98d7c532c87cc0169aef73e93788194a643ef9ddeddeae0"
is_byok
false
latency
206
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-1773074238-0hXBTb80ZOKwCQZ32SBY"
upstream_id
"resp_0eb98d7c532c87cc0169aef73e93788194a643ef9ddeddeae0"
total_cost
0.02754
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
99.74%
Matches word count
n/a
neededClean
false
words
196
99.95%
Dialogue to Total Word Ratio
Ratio: 48.48%, Deviation: 1.52%
neededClean
false
wordsTotal
198
wordsDialogue
96
99.8458%