NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM08917P3YHS0TJQADWYC8PQ
Run Details
Model:
GPT-5.4 Mini (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
4.3s
Cost:
$0.003605
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
60.28%
Matches word count
(details)
85.96%
Dialogue to Total Word Ratio
(details)
73.1180%
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
Molly eased the curtain back. "Tell me you brought the party map." "Map?" Bill whispered. "I brought nerves and a very official-looking tie." "That’s not a tie, that’s ribbon from a gift bag." "Exactly. Camouflage." "Bill, we're here to rob a birthday party, not audition for it." "Same thing if the cake's good." "Focus. Presents." "Right. Big haul: trains, dolls, a suspiciously heavy box, and something that squeaks." "If one of those is a puppy, I quit." "Nobody said puppy." "You had better not be thinking what I think you're thinking." "I'm thinking we walk in, blend with the parents, and load the gifts into the stroller." "We don't have a stroller." "Then why did I spend thirty minutes folding one?" "Bill!" "Relax. I also brought balloons." "Why?" "Because thieves who commit crimes should arrive with spirit." "That's the worst sentence you've ever said." "Yet here we are." "On three, we enter, smile, and don't mention the getaway van in clown paint." "Too late?" "Bill." "Fine. But if anyone asks, we're entertainment." "Molly." "Yes?" "If I get cake, I want the blue frosting." "You're impossible."
created_at
"2026-03-18T10:36:18.051Z"
3/18/2026, 10:36:18 AM
model
"openai/gpt-5.4-mini-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
3919
moderation_latency
(null)
generation_time
4130
tokens_prompt
77
tokens_completion
328
native_tokens_prompt
66
native_tokens_completion
790
native_tokens_completion_images
0
native_tokens_reasoning
516
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
0.0036045
router
(null)
provider_responses
0
endpoint_id
"9ee065b2-3d1c-43bc-bdd7-28af3b148282"
id
"resp_07a8608b808a85e20169ba80221a3c8190bb3fe4f430b029cc"
is_byok
false
latency
164
model_permaslug
"openai/gpt-5.4-mini-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-1773830178-0AHa2VPUrHgc1JhQYz1G"
upstream_id
"resp_07a8608b808a85e20169ba80221a3c8190bb3fe4f430b029cc"
total_cost
0.0036045
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
60.28%
Matches word count
n/a
neededClean
false
words
185
85.96%
Dialogue to Total Word Ratio
Ratio: 96.24%, Deviation: 6.24%
neededClean
false
wordsTotal
186
wordsDialogue
179
73.1180%