NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-06-01KK1HK06FDBSZ6S7WA1TEKCA4
Run Details
Model:
GPT-5.4 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
8.0s
Cost:
$0.004500
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
43.38%
Matches word count
(details)
99.53%
Dialogue to Total Word Ratio
(details)
71.4517%
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 and Bill crouched behind the inflatable castle in the church hall car park, dressed like budget magicians and smelling faintly of balloon rubber and panic. Through the windows, a dozen children in paper crowns orbited a table groaning under cakes, fizzy drinks, and a mountain of brightly wrapped presents. A clown on stilts folded a balloon sword and looked deeply unhappy, as if he too had made poor life choices. Bill peered over a hedge, notebook trembling in his hand. “We go in cheerful,” he whispered. “Fast hands, faster feet.” Molly adjusted her sequined cape and studied the battlefield. The parents were trapped by lukewarm coffee near the raffle, the children were hypnotized by a woman preparing to release a rabbit from somewhere anatomically implausible, and the gift pile sat gloriously unattended beneath a banner reading HAPPY SIXTH BIRTHDAY, OLIVER. “Take the blue bags first,” she said. “They look expensive.” A bubble machine erupted near the doorway, vomiting soap across the paving slabs. Bill stepped back, skidded, and windmilled into a cardboard pirate. The pirate collapsed with tragic dignity. Inside, the children cheered, assuming this was entertainment. Molly smiled coldly, hoisted an empty laundry sack, and marched toward the door with the confidence of a woman who had once robbed a garden centre in broad daylight.
created_at
"2026-03-06T12:22:34.456Z"
3/6/2026, 12:22:34 PM
model
"openai/gpt-5.4-20260305"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
354
moderation_latency
(null)
generation_time
7908
tokens_prompt
77
tokens_completion
356
native_tokens_prompt
66
native_tokens_completion
289
native_tokens_completion_images
0
native_tokens_reasoning
0
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
0.0045
router
(null)
provider_responses
0
endpoint_id
"9ff5625c-403f-4d7f-b895-58ac7295062c"
id
"resp_0b6d91535bc524ca0169aac70a8a708190868e54aa7162d490"
is_byok
false
latency
154
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-1772799754-dRYvemhO2ZRpifzjmAX3"
upstream_id
"resp_0b6d91535bc524ca0169aac70a8a708190868e54aa7162d490"
total_cost
0.0045
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
43.38%
Matches word count
n/a
neededClean
false
words
217
99.53%
Dialogue to Total Word Ratio
Ratio: 7.37%, Deviation: 2.63%
neededClean
false
wordsTotal
217
wordsDialogue
16
71.4517%