NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-09-01KK9Q93YPVQYCPDP3REJE4F43
Run Details
Model:
GPT-5.4 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
44.0s
Cost:
$0.041520
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
99.38%
Matches word count
(details)
100.00%
Dialogue to Total Word Ratio
(details)
99.6884%
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 behind the inflatable castle like a panther in borrowed sequins, studying the suburban battlefield. Beyond the hedge, balloons bobbed above a lawn swarming with sugar-drunk children, distracted parents, and one clown who looked morally flexible. Bill, wedged beside her in a van labeled TINY PONY PETTING ZOO, unfolded the party map with the solemnity of a general and the sneeze of a man allergic to ambition. The presents sat under a gazebo, stacked in bright paper towers, each one practically winking at crime. "Smile and wave," Molly whispered. "You distract the clown. I grab presents. No heroics, Bill." "I hate heroics," Bill whispered back. "I prefer planning, quietly." A sprinkler clicked on and soaked his shoe. He glared at it as though betrayed by plumbing itself. Molly adjusted her tiara, checked the cake knife in her boot, and timed the chaos: piƱata in seven minutes, candles in twelve, parental attention span already collapsing. Somewhere, a child screamed with the high, pure note of a teakettle discovering betrayal. Molly grinned. Every great robbery needed cover. Nothing on earth provided better cover than a birthday party. Bill straightened his fake pony tail and prayed the universe respected audacity disguised as entertainment on a Tuesday afternoon.
created_at
"2026-03-09T16:35:57.538Z"
3/9/2026, 4:35:57 PM
model
"openai/gpt-5.4-20260305"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
41687
moderation_latency
(null)
generation_time
43883
tokens_prompt
77
tokens_completion
330
native_tokens_prompt
66
native_tokens_completion
2757
native_tokens_completion_images
0
native_tokens_reasoning
2480
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
0.04152
router
(null)
provider_responses
0
endpoint_id
"9ff5625c-403f-4d7f-b895-58ac7295062c"
id
"resp_06904d03e4dbb2300169aef6ed9eb08197a5357ad960d0f825"
is_byok
false
latency
158
model_permaslug
"openai/gpt-5.4-20260305"
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-1773074157-NxemlCsFJcwrTC9OEB9p"
upstream_id
"resp_06904d03e4dbb2300169aef6ed9eb08197a5357ad960d0f825"
total_cost
0.04152
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
100.00%
Dialogue to Total Word Ratio
Ratio: 9.71%, Deviation: 0.29%
neededClean
false
wordsTotal
206
wordsDialogue
20
99.6884%