NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-09-01KK9QKT6C5HH7JDBNE9XHYWDY
Run Details
Model:
GPT-5.4 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
50.5s
Cost:
$0.039015
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
99.65%
Dialogue to Total Word Ratio
(details)
99.8237%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 10% 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 like a burglarious flamingo, one boot sunk in a flowerpot, studying the bungalow blazing with pastel balloons. Inside, shrieks of birthday joy rose and fell like a fire alarm played on kazoo. Bill unfolded a hand-drawn map of the garden, the paper greasy from sausage rolls and bad decisions. He had labeled the trampoline "north wall" and the clown "probable guard dog." "Front gate, cake table, presents. In, out, vanish," Molly whispered. Bill adjusted his paper crown disguise. "What about the magician?" "Smile, clap, pinch his wand, and keep moving." They watched a pirate in face paint herd children toward musical statues, leaving the loot stacked beside a cardboard castle. Bright bags glittered there, stuffed with plastic dinosaurs, glitter pens, and enough batteries to power a small coup. Bill's eyes shone with professional admiration. The plan was absurd, which meant it suited them perfectly. Molly would pose as an exhausted aunt. Bill, carrying a tray of lemonade nobody had ordered, would drift straight to the presents. By the time the candles were lit, the mountain would be gone, and the children would blame the magician. Somewhere, a balloon popped like a starting pistol that afternoon.
created_at
"2026-03-09T16:41:47.989Z"
3/9/2026, 4:41:47 PM
model
"openai/gpt-5.4-20260305"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
46609
moderation_latency
(null)
generation_time
50472
tokens_prompt
77
tokens_completion
452
native_tokens_prompt
66
native_tokens_completion
2590
native_tokens_completion_images
0
native_tokens_reasoning
2326
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
0.039015
router
(null)
provider_responses
0
endpoint_id
"9ff5625c-403f-4d7f-b895-58ac7295062c"
id
"resp_09db7b60d33213ae0169aef84c0da0819590a2aed1f247153f"
is_byok
false
latency
155
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-1773074507-B9MTKgaC8b06OFZeqNpW"
upstream_id
"resp_09db7b60d33213ae0169aef84c0da0819590a2aed1f247153f"
total_cost
0.039015
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
200
99.65%
Dialogue to Total Word Ratio
Ratio: 12.44%, Deviation: 2.44%
neededClean
false
wordsTotal
201
wordsDialogue
25
99.8237%