NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM09A4PY8D85T8AJJMZVRC1Z
Run Details
Model:
GPT-5.4 Mini (Reasoning, Low)
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
2.2s
Cost:
$0.001800
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
35.00%
Matches word count
(details)
21.90%
Dialogue to Total Word Ratio
(details)
28.4523%
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 beside the bakery dumpster, studying the glittering invitation she had fished from a pocket of discarded confetti. Across the alley, Bill was trying to look dangerous while wrestling a borrowed clown shoe onto his left foot. The shoe squeaked with every movement, like it was laughing at him personally. “Birthday party, second floor, blue balloons,” Molly whispered. “Easy.” “Nothing is easy,” Bill whispered back, adjusting his stolen bow tie. “Especially not children. They move like swarms.” She smirked and unfolded the floor plan she had drawn on the back of a placemat. The building was a community hall, all creaky stairs, paper lanterns, and one overworked volunteer at the door. Inside, there would be presents stacked under a cardboard castle, waiting to be admired, opened, and, if the universe was kind, removed. Bill peered at the plan. “You sure this is moral?” Molly tapped the page. “We are thieves, Bill. We sell morality by the ounce.” He nodded, though his expression suggested he’d rather buy it in bulk. Somewhere inside, a child shrieked with delight, followed by the rustle of wrapping paper. Molly inhaled, grinned, and tucked a flashlight into her sleeve. “Remember,” she said, “we leave with gifts, not guilt.” Bill straightened his bow tie and grinned back. “Then let’s make a very merry exit.”
created_at
"2026-03-18T10:54:22.955Z"
3/18/2026, 10:54:22 AM
model
"openai/gpt-5.4-mini-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
892
moderation_latency
(null)
generation_time
1878
tokens_prompt
77
tokens_completion
364
native_tokens_prompt
66
native_tokens_completion
389
native_tokens_completion_images
0
native_tokens_reasoning
84
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
0.0018
router
(null)
provider_responses
0
endpoint_id
"9ee065b2-3d1c-43bc-bdd7-28af3b148282"
id
"resp_048e2d75aaf5cb3d0169ba845f03bc8190a7daac33d4aa7acb"
is_byok
false
latency
177
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-1773831262-0GCIYSFtU1vixAk02FfV"
upstream_id
"resp_048e2d75aaf5cb3d0169ba845f03bc8190a7daac33d4aa7acb"
total_cost
0.0018
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
35.00%
Matches word count
n/a
neededClean
false
words
218
21.90%
Dialogue to Total Word Ratio
Ratio: 21.10%, Deviation: 11.10%
neededClean
false
wordsTotal
218
wordsDialogue
46
28.4523%