NC Bench
Overview
About
Tests
Models
Model
Claude Sonnet 4.5
Test
Dialogue tags
Scenario
Write 500 words with 30% dialogue
Run ID
2026-02-08-01KGZ0BSH4QDWD93H336N1KY0P
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 30% 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 hedgerow, peering through her binoculars at the suburban house across the street. Colorful balloons bobbed from the mailbox, and a banner proclaiming "Happy 8th Birthday Emily!" stretched across the front porch. Inside, she could see approximately fifteen screaming children bouncing around like caffeinated pinballs. Bill shuffled up beside her, breathing heavily from the short walk from their van. "This is it?" Bill wheezed. "This is the big score you've been talking about?" "Trust me," Molly whispered. "I saw the guest list. These kids are loaded." She adjusted her position, nearly losing her balance. Bill grabbed her elbow to steady her, and they both froze as a soccer ball sailed over their heads, landing somewhere in the bushes behind them. "The Richards family just bought their kid the new HyperStation 6. Those things are going for three grand on the resale market," Molly continued. Bill pulled out a crumpled piece of paper covered in chocolate fingerprints. Their handwritten plan looked more like abstract art than a proper heist strategy. "So we're really doing this? Breaking into a children's birthday party?" "We're liberating presents from people who have too much," Molly corrected him. "It's practically a public service." Inside the house, a magician appeared to be pulling a rabbit from a hat while the children shrieked with delight. Molly noted the back door opening periodically as parents stepped out to take phone calls or escape the chaos. "Okay, so what's the entry point?" Bill asked. Molly pointed to the side gate leading to the backyard. A clown was currently setting up balloon animals near a picnic table absolutely drowning in wrapped packages. "We go in during the piñata. Everyone will be distracted. You create a diversion at the front, maybe pretend to be delivering more pizza or something. I slip in through the side, grab as many presents as I can carry, and we're gone in ninety seconds." "What kind of diversion?" "I don't know, be creative. Dance. Sing. Tell them you're with animal control and there's been reports of an escaped peacock in the neighborhood." Bill stared at her blankly. "A peacock?" "Work with me here." They watched as Emily's mother emerged from the house carrying an enormous piñata shaped like a unicorn. The children immediately abandoned all other activities and swarmed toward it like tiny, sugar-crazed zombies. "That's our cue in five minutes," Molly said, checking her watch. Bill stood up, brushing leaves from his pants. He looked at the house, then back at Molly, then at the house again. His face had taken on a slightly greenish tinge. "I need you to know that this is absolutely the lowest point of my criminal career," he said. "Yeah, well, we all have to start somewhere." Molly grinned and pulled on a pair of black gloves. "Now go find yourself a peacock story that's believable."
created_at
"2026-02-08T16:11:53.945081+00:00"
Sunday, February 8th, 2026 at 4:11:53 PM GMT+00:00
model
"anthropic/claude-4.5-sonnet-20250929"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1531
moderation_latency
188
generation_time
13083
tokens_prompt
61
tokens_completion
614
native_tokens_prompt
75
native_tokens_completion
713
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
0.01092
router
(null)
provider_responses
0
id
"ebd887c2-5732-4878-84ea-c7d6f5cf4206"
status
200
is_byok
false
latency
1528
endpoint_id
"d8bee848-f020-454a-b2be-4a7ce868f99f"
provider_name
"Amazon Bedrock"
model_permaslug
"anthropic/claude-4.5-sonnet-20250929"
api_type
"completions"
id
"gen-1770567099-2j616JxL3nxPBzfrky6J"
upstream_id
"ebd887c2-5732-4878-84ea-c7d6f5cf4206"
total_cost
0.01092
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Amazon Bedrock"
Evaluation details
Result
Evaluator
Details
Meta Data
14.3014%
Matches word count
n/a
neededClean
false
words
479
0.1276%
Dialogue to Total Word Ratio
Ratio: 41.54%, Deviation: 11.54%
neededClean
false
wordsTotal
479
wordsDialogue
199
7.2145%