NC Bench
Overview
About
Tests
Models
Model
Claude Sonnet 4.5
Test
Dialogue tags
Scenario
Write 500 words with 30% dialogue
Run ID
2026-02-08-01KGYZ72AV115RRFV2JE0VM6Y7
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 back against the inflatable bounce house, its rubbery surface squeaking against her leather jacket. Through the chain-link fence, she could see the mountain of wrapped presents stacked beneath a pink and purple unicorn banner. Bill crouched beside her, consulting what appeared to be a hand-drawn map on a napkin. The suburban backyard was a fortress of festivities. Balloons bobbed from every available surface, and a clown was performing increasingly dangerous juggling tricks near the snack table. At least twenty screaming children ran in chaotic patterns across the lawn, hopped up on what Molly assumed was enough sugar to fuel a rocket launch. Bill adjusted his sunglasses and studied the scene with the intensity of a general surveying a battlefield. "The presents are exposed," Bill whispered. "We go in during the piñata." "That's your plan? Just waltz in during the piñata?" "Maximum distraction. Everyone'll be watching the candy scramble." Molly shook her head. The whole operation felt beneath her skill set. Last month she had lifted a Rembrandt from a private collection. Now she was casing a seven-year-old's birthday party because Bill's gambling debts had gotten creative. "Remind me again why we're stealing from children," Molly said. "One of those boxes contains a vintage mint-condition Dragon Master action figure. The collector's paying fifteen grand." "Fifteen thousand dollars?" "Limited edition. Only fifty made. The birthday girl's weird uncle bought it, didn't realize what he had." A mother emerged from the house carrying an enormous sheet cake, and the children converged like a swarm of locusts. Bill tensed, ready to move, but Molly grabbed his arm. "Wait. Which present is it?" Bill consulted his napkin again, squinting at his own handwriting. "Blue wrapping paper. Silver bow." Molly scanned the present pile. Her heart sank. "Bill. There are eleven blue presents with silver bows." Bill's confident expression crumbled. He looked at the pile, then at his napkin, then back at the pile. His mouth opened and closed like a confused goldfish. "We'll take them all," Bill declared. "Are you insane? We can't carry eleven presents. Someone will definitely notice." Before Bill could respond, the piñata emerged, a massive unicorn suspended from the oak tree. The children screamed with primal delight. A blindfolded boy was handed a baseball bat that seemed far too large for the task. "Now or never," Bill said, and before Molly could stop him, he was over the fence. She watched in horror as Bill sprinted across the lawn in a low crouch, looking like a deranged crab. He reached the present table just as the piñata exploded, and candy rained down upon the shrieking masses. Molly sighed deeply and checked her watch. She gave Bill forty-five seconds before this whole thing went sideways. She was already planning their escape route through the neighbor's yard.
created_at
"2026-02-08T15:51:52.268751+00:00"
Sunday, February 8th, 2026 at 3:51:52 PM GMT+00:00
model
"anthropic/claude-4.5-sonnet-20250929"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1245
moderation_latency
373
generation_time
14572
tokens_prompt
61
tokens_completion
613
native_tokens_prompt
75
native_tokens_completion
697
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
0.01068
router
(null)
provider_responses
0
id
"c659e254-256d-42cd-82e6-7a9d41d62db1"
status
200
is_byok
false
latency
1245
endpoint_id
"d8bee848-f020-454a-b2be-4a7ce868f99f"
provider_name
"Amazon Bedrock"
model_permaslug
"anthropic/claude-4.5-sonnet-20250929"
api_type
"completions"
id
"gen-1770565896-sqVvBbIZF5SUv8PcpZVi"
upstream_id
"c659e254-256d-42cd-82e6-7a9d41d62db1"
total_cost
0.01068
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Amazon Bedrock"
Evaluation details
Result
Evaluator
Details
Meta Data
0.4920%
Matches word count
n/a
neededClean
false
words
473
11.8380%
Dialogue to Total Word Ratio
Ratio: 23.47%, Deviation: 6.53%
neededClean
false
wordsTotal
473
wordsDialogue
111
6.1650%