NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM098GNDNXZAE6FKHKH9VSSH
Run Details
Model:
GPT-5.4 Mini (Reasoning, Low)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
2.7s
Cost:
$0.001683
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
99.38%
Matches word count
(details)
32.59%
Dialogue to Total Word Ratio
(details)
65.9848%
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 you’ve got a plan,” Bill whispered, peering at the row of balloons bobbing over the garden fence. “I’ve got a plan, a ladder, and a deeply concerning amount of clown makeup,” Molly said. “We go in as entertainers.” “You mean thieves.” “I mean entertainers with portable pockets.” Bill frowned. “What about the birthday kid?” “What about him?” “He’s six.” “Perfect. I’ve met six-year-olds with better security systems than the museum.” Bill adjusted his cap. “You’re sure the presents are worth it?” “Bill, this is a children’s party. Those boxes are like treasure chests with ribbons.” “And the cake?” “If it’s chocolate, yes. If it’s fondant, no one deserves that.” A shriek of laughter burst through the fence. Molly grinned. “Hear that? Chaos. The adults are distracted, the kids are sticky, and the presents are unattended.” “You make crime sound festive.” “It is festive. We’re basically seasonal.” Bill leaned closer. “And if someone asks who we are?” “Mister Wiggles and Aunt Sparkle.” “You can’t be Aunt Sparkle.” “Why not?” “Because you look like you’d bite a policeman.” Molly sighed. “Fine. You be Aunt Sparkle. I’ll be Mister Wiggles.” Bill straightened his tie. “I hate this already.” “Good,” said Molly. “That means you’re ready.”
created_at
"2026-03-18T10:53:29.657Z"
3/18/2026, 10:53:29 AM
model
"openai/gpt-5.4-mini-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1482
moderation_latency
(null)
generation_time
2436
tokens_prompt
77
tokens_completion
403
native_tokens_prompt
66
native_tokens_completion
363
native_tokens_completion_images
0
native_tokens_reasoning
56
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
0.001683
router
(null)
provider_responses
0
endpoint_id
"9ee065b2-3d1c-43bc-bdd7-28af3b148282"
id
"resp_0e356bb359e226230169ba8429bc7481908d70d49eb2d7672d"
is_byok
false
latency
212
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-1773831209-ykZVUTi9hd5IC4xPVHhZ"
upstream_id
"resp_0e356bb359e226230169ba8429bc7481908d70d49eb2d7672d"
total_cost
0.001683
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
99.38%
Matches word count
n/a
neededClean
false
words
205
32.59%
Dialogue to Total Word Ratio
Ratio: 79.71%, Deviation: 10.29%
neededClean
false
wordsTotal
207
wordsDialogue
165
65.9848%