NC Bench
Overview
About
Tests
Models
Model
Claude Sonnet 4.5
Test
Dialogue tags
Scenario
Write 500 words with 70% dialogue
Run ID
2026-02-08-01KGYXG196Y3B9QVVM8JWMYQYW
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 70% 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 pressed her back against the fence and peered through a knothole at the bouncy castle. "Bill, we've got approximately forty-seven minutes before the piñata breaks and all hell breaks loose." "Forty-seven? That's oddly specific." Bill crouched beside her, adjusting his ridiculous mustache disguise. "I called the party planner. Pretended to be the birthday boy's aunt." "You're terrifying." Bill squinted through the fence. "Okay, I count six adults, maybe twenty kids. What's our entry point?" "The gift table is positioned near the back door. Classic rookie mistake by the parents." "We can't just waltz in there, Molly. It's broad daylight. There are children." "That's exactly why we can. Who's going to suspect two party entertainers?" Molly pulled out two t-shirts from her bag. One read 'Captain Fantastic' and the other said 'Princess Sparkle McGuire.' "Absolutely not. I'm not wearing that." "You want the Vermeer money or not?" Bill sighed. "Remind me again why we think this kid has sixteen grand in gift cards?" "His father is Marcus Chen. Tech billionaire. Throws money around like confetti. The rumor is every relative just gives gift cards because what do you buy a five-year-old who has everything?" "This is insane. Even for us." "Says the man who once stole a garden gnome thinking it was a Fabergé egg." "That was one time and the lighting was poor." Bill grabbed the Captain Fantastic shirt. "Fine. What's the play?" "We go in, you do your terrible magic tricks to distract everyone while I swap the gift bags with our decoys." "My magic tricks aren't terrible." "Bill, you once made a rabbit disappear permanently. We still don't know where Mr. Whiskers went." "He's living his best life somewhere, I'm sure." Bill pulled the shirt over his head. "What's in the decoy bags?" "IOUs and inflatable dinosaurs. I got them in bulk." "You think IOUs will satisfy a tech billionaire's family?" "By the time they figure it out, we'll be in Cabo." Bill stood up and adjusted his mustache again. "This is officially the weirdest job we've ever pulled." "Excuse me, did you forget about the time we infiltrated that dog show?" "We don't talk about the dog show, Molly." "You looked good with the poodle cut." "I said we don't talk about it." Bill took a deep breath. "Okay, Captain Fantastic is ready. Let's go steal some birthday presents from a kindergartener. I can't believe this is my life." "Think of the Vermeer." "I'm thinking of the Vermeer." Molly opened the gate and painted on her biggest smile. "Remember, you're fantastic and I'm sparkly." "If we get arrested, I'm blaming you entirely." "Get in line. My parole officer has dibs."
created_at
"2026-02-08T15:21:49.286191+00:00"
Sunday, February 8th, 2026 at 3:21:49 PM GMT+00:00
model
"anthropic/claude-4.5-sonnet-20250929"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1353
moderation_latency
175
generation_time
15500
tokens_prompt
61
tokens_completion
620
native_tokens_prompt
75
native_tokens_completion
718
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
0.010995
router
(null)
provider_responses
0
id
"aaa9c218-adb6-4296-b60a-3c7d1abafb3c"
status
200
is_byok
false
latency
1352
endpoint_id
"d8bee848-f020-454a-b2be-4a7ce868f99f"
provider_name
"Amazon Bedrock"
model_permaslug
"anthropic/claude-4.5-sonnet-20250929"
api_type
"completions"
id
"gen-1770564093-JvbGxsuxZdVJr0Fzkzet"
upstream_id
"aaa9c218-adb6-4296-b60a-3c7d1abafb3c"
total_cost
0.010995
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Amazon Bedrock"
Evaluation details
Result
Evaluator
Details
Meta Data
0.0000%
Matches word count
n/a
neededClean
false
words
448
0.0362%
Dialogue to Total Word Ratio
Ratio: 82.59%, Deviation: 12.59%
neededClean
false
wordsTotal
448
wordsDialogue
370
0.0181%