NC Bench
Overview
About
Tests
Models
Model
Claude Sonnet 4.5
Test
Dialogue tags
Scenario
Write 500 words with 50% dialogue
Run ID
2026-02-08-01KGYZGTQXVDB10QBRV3BBGPVM
Model input
User/Human
You are a author of comedic heist books. Please write a 500 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
The van lurched to a stop three houses down from the target. Molly peered through binoculars at the inflatable bounce castle dominating the front yard. "Seventeen children confirmed," she muttered. "Plus at least four adults." Bill unwrapped a protein bar. "What's the birthday kid's name again?" "Sophia. She's turning seven." Molly lowered the binoculars. "The present table is set up near the back fence. I counted at least twenty-five wrapped boxes." "Twenty-five? For one kid?" Bill shook his head. "That's excessive." "Her parents are loaded. Tech money." Molly pulled out a hand-drawn map. "Here's the plan. You go in through the side gate during the piñata distraction." "Wait, there's a piñata?" "Of course there's a piñata. It's a children's party." Bill crumpled his wrapper. "What if there's no piñata?" "There will be. I hacked the mother's email. She ordered a unicorn piñata from Party Palace." "You hacked her email for a piñata confirmation?" Molly glared at him. "I'm thorough. Do you want to hear the plan or not?" "Fine, continue." "When the piñata comes out, all eyes will be on the blindfolded kid with the stick. You slip in, grab as many presents as possible, load them into the garden wagon by the fence." Bill leaned forward. "They have a garden wagon?" "Red Radio Flyer. I saw the dad using it yesterday to haul mulch." "This is actually happening. We're robbing a seven-year-old's birthday party." "We're not robbing her. We're robbing her rich parents who think material possessions equal love." Molly checked her watch. "The piñata should start in fifteen minutes." "What's my exit strategy?" "Same way you came in. Side gate. I'll be waiting with the van running." Bill pulled on a baseball cap. "What if someone stops me?" "Tell them you're with the pony." "What pony?" "The pony arriving at three o'clock. Also confirmed via hacked email." "There's a pony?" Bill's eyes widened. "An actual pony?" "Her name is Butterscotch. She does birthday parties on weekends." "This family has everything." Molly handed him a neon safety vest. "Wear this. Everyone ignores people in safety vests. You'll blend right in." "As what? The birthday safety inspector?" "As someone who belongs there. That's the key to any successful heist. Confidence." Bill slipped on the vest. "I still think this is crazy." "You said that about the nursing home job." "Because it was crazy! We stole seventy-five birthday cards from a ninety-year-old woman." "And made three thousand dollars. Those cards had cash inside." Molly started the engine. "Now focus. We're professionals." "We're stealing from children." "We're redistributing wealth. Besides, we're donating half to that animal shelter." "The one you're banned from?" "I'm not banned. I'm on probation." Molly shifted into gear. "Now get ready. Operation Tiny Tyrant begins in ten minutes."
created_at
"2026-02-08T15:57:13.363594+00:00"
Sunday, February 8th, 2026 at 3:57:13 PM GMT+00:00
model
"anthropic/claude-4.5-sonnet-20250929"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1554
moderation_latency
194
generation_time
16174
tokens_prompt
61
tokens_completion
664
native_tokens_prompt
75
native_tokens_completion
785
native_tokens_completion_images
(null)
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
"stop"
usage
0.012
router
(null)
provider_responses
0
id
"18b15766-95bb-4a71-8314-4862eb527ba4"
status
200
is_byok
false
latency
1554
endpoint_id
"d8bee848-f020-454a-b2be-4a7ce868f99f"
provider_name
"Amazon Bedrock"
model_permaslug
"anthropic/claude-4.5-sonnet-20250929"
api_type
"completions"
id
"gen-1770566216-YRlDGYSmrhMg03KSbzUI"
upstream_id
"18b15766-95bb-4a71-8314-4862eb527ba4"
total_cost
0.012
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Amazon Bedrock"
Evaluation details
Result
Evaluator
Details
Meta Data
0.0028%
Matches word count
n/a
neededClean
false
words
468
0.0000%
Dialogue to Total Word Ratio
Ratio: 80.56%, Deviation: 30.56%
neededClean
false
wordsTotal
468
wordsDialogue
377
0.0014%