NC Bench
Overview
About
Tests
Models
Model
Claude Sonnet 4.5
Test
Dialogue tags
Scenario
Write 500 words with 30% dialogue
Run ID
2026-02-08-01KGZ0ABZRXR984MD86DHKM18T
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 pressed her face against the fence, peering through a knothole at the chaos beyond. Balloons bobbed in the breeze. A piñata shaped like a unicorn swayed ominously from an oak tree. Children shrieked with sugar-fueled abandon. What am I doing with my life, she wondered. Bill crouched beside her, adjusting his unnecessarily elaborate mustache disguise. The thing looked like a deceased caterpillar had taken up residence on his upper lip. You really think this is our best option? Bill nodded enthusiastically, his fake mustache sliding slightly to the left. The birthday girl's parents are loaded. Those presents are probably worth thousands. Plus, I heard they ordered that new HyperDrone 5000. That thing retails for eight hundred bucks. Molly sighed and checked her watch. The party had been raging for forty-five minutes. In her experience, that meant the parents would be sufficiently distracted by the impending cake ceremony. She'd planned seventeen heists in her career, each more successful than the last. None had involved stealing from children. Her partner stood and brushed grass from his knees. He'd insisted on wearing what he called their infiltration outfits, which consisted of painfully bright Hawaiian shirts and cargo shorts. According to Bill's logic, they'd blend right in as quirky uncles. Remember the plan? Bill recited it like a bored student. I create a distraction with the backup piñata. You slip around through the garage. Grab the present table haul. We meet at the extraction point behind the tool shed. The backup piñata sat in their van, stuffed not with candy but with live crickets. Bill had been unreasonably proud of that particular innovation. Molly scaled the fence with practiced ease, landing softly in a flower bed. Bill followed with significantly less grace, toppling into a hydrangea bush. A nearby toddler watched them with suspicious eyes while gnawing on a juice box. They separated, moving through the yard with what Bill probably thought was stealth. Molly watched him lumber toward the piñata, cricket-filled backup clutched under one arm. The man moved like a bear in board shorts. She slipped through the open garage door. Beyond it lay the kitchen, and through the kitchen window, she spotted the glorious present table. It practically sparkled in the afternoon sun. Gift wrap in every color imaginable, bows the size of dinner plates, bags overflowing with tissue paper. The sound of shattering cricket-piñata reached her ears. Children screamed. Parents shouted. Someone yelled about locusts, which seemed dramatic. Molly darted forward, grabbed the loaded wagon Bill had positioned earlier, and started loading presents with the efficiency of a seasoned professional. Within ninety seconds, she had everything. She wheeled her haul toward the tool shed. Bill appeared moments later, covered in glitter and what appeared to be frosting. Mission accomplished? Mission accomplished. Now let's get out of here before someone calls the cops. They ran.
created_at
"2026-02-08T16:11:09.619921+00:00"
Sunday, February 8th, 2026 at 4:11:09 PM GMT+00:00
model
"anthropic/claude-4.5-sonnet-20250929"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1355
moderation_latency
238
generation_time
15337
tokens_prompt
61
tokens_completion
612
native_tokens_prompt
75
native_tokens_completion
692
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
0.010605
router
(null)
provider_responses
0
id
"c27a5e00-c18f-4134-b65d-44bf24604622"
status
200
is_byok
false
latency
1354
endpoint_id
"d8bee848-f020-454a-b2be-4a7ce868f99f"
provider_name
"Amazon Bedrock"
model_permaslug
"anthropic/claude-4.5-sonnet-20250929"
api_type
"completions"
id
"gen-1770567053-xN9wNtbvLAhBYCEyYBTE"
upstream_id
"c27a5e00-c18f-4134-b65d-44bf24604622"
total_cost
0.010605
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
0.0000%
Dialogue to Total Word Ratio
Ratio: 0.00%, Deviation: 30.00%
neededClean
false
wordsTotal
479
wordsDialogue
0
7.1507%