NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM07JR7JV7SH3CV7XCAN0MEF
Run Details
Model:
GPT-5.4 Mini (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
5.9s
Cost:
$0.005801
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
49.9995%
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, tell me again why we're robbing a party full of cake and tiny saxophone music?" Bill whispered. "Because, Bill, the birthday boy has a pirate ship for gifts, a drone, and something wrapped so badly it might be a telescope." "That sounds expensive." "Exactly." "I prefer jobs with alarms, vaults, and fewer clowns." "There is one clown." "That is one clown too many." Molly adjusted her black gloves. "Listen. We stroll in as the entertainment." "We're thieves, not jugglers." "We can juggle." "You can juggle. I can panic rhythmically." "Fine. You be the magician's assistant." "We don't have a magician." "Then be the magician's disappointment." Bill sighed. "And the presents?" "While the parents sing, we load them into the catering trolley." "Won't anyone notice?" "Molly, it's a children's party. Adults are always too busy filming and pretending to enjoy sugar." Bill grinned despite himself. "And if a child asks why the elf is carrying a mountain of toys?" "I'll say you're taking them to Santa's emergency garage." "Brilliant." "Ready?" "No." "Good. That's the spirit." He squared his shoulders. "For the record, if we get caught, I'm blaming the clown." "Fair," Molly said, and together they pushed open the glittery door.
created_at
"2026-03-18T10:24:07.934Z"
3/18/2026, 10:24:07 AM
model
"openai/gpt-5.4-mini-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
5192
moderation_latency
(null)
generation_time
5735
tokens_prompt
77
tokens_completion
329
native_tokens_prompt
66
native_tokens_completion
1278
native_tokens_completion_images
0
native_tokens_reasoning
984
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
0.0058005
router
(null)
provider_responses
0
endpoint_id
"9ee065b2-3d1c-43bc-bdd7-28af3b148282"
id
"resp_0abcba669c035b560169ba7d47fcf48195858436c9605301e1"
is_byok
false
latency
167
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-1773829447-rmmuor7s6ehEBdTCZkBP"
upstream_id
"resp_0abcba669c035b560169ba7d47fcf48195858436c9605301e1"
total_cost
0.0058005
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
0.00%
Dialogue to Total Word Ratio
Ratio: 68.16%, Deviation: 21.84%
neededClean
false
wordsTotal
201
wordsDialogue
137
49.9995%