NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM07TQ8QKQYWJ951QXQ5MMFR
Run Details
Model:
GPT-5.4 Nano (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
14.9s
Cost:
$0.004056
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
39.73%
Dialogue to Total Word Ratio
(details)
69.8635%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 90% 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, are you sure this is the right house?" Molly whispered. "Heck yes. Look for the inflatable dinosaur. No normal criminal has one." Molly peered at the gate. "We’re not criminals, we’re… professional gift aspirants." Bill adjusted his beanie. "Professional gift aspirants who steal all the presents." "Shh. The kids are practicing 'Jingle Bells' as burglar alarms." Bill handed her a sack. "That’s why we wear headphones." Molly laughed. "Headphones? To block screaming, or to muffle your monologues?" "Both. My monologue is top-tier." "Okay, plan: you go through the dog door." Molly blinked. "There’s a sign that says, 'No thieves, only treats.'" Bill shrugged. "We’ll bring treats. Mostly." Molly: "What if the dog recognizes you?" Bill: "I trained with a chew toy. I’m a hero." Molly: "Fine, I’ll distract the parents with your 'innocent face.'" Bill: "It’s not innocent, it’s… collectible." A car rolled by outside. Molly said, "When the lights turn off for cake, we grab the boxes, then sprint." Bill nodded. "And if someone asks, we’re Santa’s accounting team." Molly sighed. "Santa doesn’t need accounting." Bill: "He will after we steal the spreadsheets—uh, presents." Molly: "Let’s do it. For glory and wrapping paper." Bill: "For glory and witnesses!"
created_at
"2026-03-18T10:28:29.093Z"
3/18/2026, 10:28:29 AM
model
"openai/gpt-5.4-nano-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
13643
moderation_latency
(null)
generation_time
14683
tokens_prompt
77
tokens_completion
354
native_tokens_prompt
66
native_tokens_completion
3234
native_tokens_completion_images
0
native_tokens_reasoning
2885
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
0.0040557
router
(null)
provider_responses
0
endpoint_id
"0c835f2e-c18d-4e8c-b245-e1e3bd08b97f"
id
"resp_004b8aca7fd494b90169ba7e4d2c088195992e7ccef7f45445"
is_byok
false
latency
193
model_permaslug
"openai/gpt-5.4-nano-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-1773829709-0nEG8Fjsiex7tTeeZMyz"
upstream_id
"resp_004b8aca7fd494b90169ba7e4d2c088195992e7ccef7f45445"
total_cost
0.0040557
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
201
39.73%
Dialogue to Total Word Ratio
Ratio: 80.20%, Deviation: 9.80%
neededClean
false
wordsTotal
202
wordsDialogue
162
69.8635%