NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-09-01KKAC9JZJ3WR8TYW4CAYT5EZW
Run Details
Model:
GPT-5.4 (Reasoning, Low)
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
24.9s
Cost:
$0.022920
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
99.22%
Dialogue to Total Word Ratio
(details)
99.6123%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 50% 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 garden shed, studying the inflatable castle as if it were a bank vault with balloons. "Security is laughable," whispered Molly. "One clown, two distracted parents, and a magician whose rabbit is clearly freelancing." Bill adjusted his black turtleneck and stared at the buffet. "Forget the gifts, I'm stealing the sausage rolls." "Focus," Molly hissed. "Presents first, pastry later. We go in as face painters." Bill frowned. "I can't paint faces. Last time I made a child look like Mussolini." "Then do tigers," said Molly. "Every amateur becomes a tiger eventually." A burst of shrieking erupted from the trampoline. Bill flinched. "What if they recognize us?" "From where?" Molly said. "That unfortunate christening job." Bill considered the rainbow-wrapped mountain beside the cake. "Fine. In, grab, out. No improvising." Molly smiled, producing two tiny party hats. "Bill, if a plan involves loot and elastic chin straps, improvising is the whole profession." Near the gate, a pirate-themed piƱata swung in the breeze like a taunting safe. "After you," Bill muttered. "Children smell fear." Molly straightened, dabbed blue paint beneath his eyes, and handed him a brush. "Excellent, then you'll pass for a sad raccoon at the county fair tonight."
created_at
"2026-03-09T22:43:13.023Z"
3/9/2026, 10:43:13 PM
model
"openai/gpt-5.4-20260305"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
22233
moderation_latency
(null)
generation_time
24801
tokens_prompt
77
tokens_completion
554
native_tokens_prompt
66
native_tokens_completion
1517
native_tokens_completion_images
0
native_tokens_reasoning
1220
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
0.02292
router
(null)
provider_responses
0
endpoint_id
"9ff5625c-403f-4d7f-b895-58ac7295062c"
id
"resp_0a59f59e716bf1220169af4d011b948196a15b38394b8d9674"
is_byok
false
latency
251
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-1773096193-vF4TLPjUDJOmZ0XpuD3M"
upstream_id
"resp_0a59f59e716bf1220169af4d011b948196a15b38394b8d9674"
total_cost
0.02292
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
200
99.22%
Dialogue to Total Word Ratio
Ratio: 52.97%, Deviation: 2.97%
neededClean
false
wordsTotal
202
wordsDialogue
107
99.6123%