NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-09-01KKAC8FSRZ7R2WW20EB3DB9D5
Run Details
Model:
GPT-5.4 (Reasoning, Low)
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
18.9s
Cost:
$0.019590
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
6.43%
Dialogue to Total Word Ratio
(details)
53.2135%
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 studied the suburban fortress from behind a hedge shaped like a startled rabbit. Balloons bobbed over the fence, and somewhere inside a clown honked with the menace of a foghorn. “Bill, focus,” she whispered. “At three o’clock the magician starts. Every parent turns toward the rabbit trick. We slip past the patio, hit the gift table, and vanish before cake.” Bill adjusted his black turtleneck, already dusted with cupcake icing. “Elegant. But what if the clown spots us?” “Then improvise. Fall down, juggle, look employable.” “I’ve never looked employable.” “Exactly. They’ll assume you’re hired entertainment.” Bill peered through the leaves. “There are at least twelve children. Sticky, unpredictable, and fueled by orange squash. I’d rather rob a bank.” “Banks have alarms. Six-year-olds have attention spans.” “And grandparents,” Bill said. “The old ones are sharp. One sniffed me at Easter.” Molly grinned. “That’s why I brought the decoy.” From her bag she produced a glittering piñata shaped like a unicorn. Bill blinked. “We’re distracting children with a larger present?” “With greed,” Molly said. “The purest security exploit.” Inside, the clown honked again. Bill sighed. “Fine. But if I’m forced to make balloon animals, I’m taking hazard pay, plus dental coverage.”
created_at
"2026-03-09T22:42:36.993Z"
3/9/2026, 10:42:36 PM
model
"openai/gpt-5.4-20260305"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
16594
moderation_latency
(null)
generation_time
18810
tokens_prompt
77
tokens_completion
861
native_tokens_prompt
66
native_tokens_completion
1295
native_tokens_completion_images
0
native_tokens_reasoning
994
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
0.01959
router
(null)
provider_responses
0
endpoint_id
"9ff5625c-403f-4d7f-b895-58ac7295062c"
id
"resp_0b7a480f601328ad0169af4cdd0fec819385304e4b63269b41"
is_byok
false
latency
166
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-1773096156-DocTciKQNVztKVKZGDJy"
upstream_id
"resp_0b7a480f601328ad0169af4cdd0fec819385304e4b63269b41"
total_cost
0.01959
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
6.43%
Dialogue to Total Word Ratio
Ratio: 62.87%, Deviation: 12.87%
neededClean
false
wordsTotal
202
wordsDialogue
127
53.2135%