NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-01-01KWEA6E27QY749W8FT19S7K5H
Run Details
Model:
Claude Sonnet 5 (Reasoning, Low)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
21.4s
Cost:
$0.011316
Provider:
Amazon Bedrock
Evaluation summary
Score
Evaluator
Details Link
97.63%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
48.8138%
Overall
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
The warehouse smelled of stale popcorn and broken dreams, which seemed fitting given that Bill had chosen it as their base of operations for reasons known only to him. Molly sat cross-legged on an overturned crate, studying the crumpled blueprint of the Hendricks' backyard like it was the Rosetta Stone. "A bounce house," she muttered. "Who puts a bounce house directly in front of the present table? It's like they're daring us." Bill paced behind her, munching on a stolen granola bar he'd lifted from a gas station three towns over. "That's the genius of it, Molly. Nobody suspects the bounce house. It's a blind spot. A soft, inflatable blind spot." "You want us to hide inside a bounce house." "I want us to hide *behind* the bounce house. There's a difference. One plan gets us thirty unwrapped presents. The other gets us arrested next to a clown named Sparkles." Molly rubbed her temples. Six years of partnership with this man, and she still hadn't figured out whether he was a criminal mastermind or a golden retriever with opposable thumbs. "And what's your plan for the actual children? You know, the eight-year-olds who will absolutely notice two grown adults stealing their birthday loot." "Cake." "Cake?" "Everyone loves cake, Molly. We show up, we're 'entertainers,' we distract with a magic trick, and while thirty sugar-crazed children are face-first in buttercream, we grab the goods and vanish like ninjas." "Ninjas who drive a rusted-out minivan with a broken taillight." "The taillight adds character." Molly stood, tossing the blueprint aside. "Fine. Say we get past the cake, past the bounce house, past Sparkles the clown who is apparently armed with balloon animals. How exactly do we carry thirty presents out of a backyard without anyone noticing?" Bill grinned, the kind of grin that had gotten them both into trouble more times than Molly cared to count. He reached behind the crate and produced two enormous black trash bags, already fashioned into makeshift capes. "We're superheroes," he said. "Captain Grab and the Snatcher." "Absolutely not." "The kids will love it! We swoop in, we do a bit, we swoop out with the presents. It's not stealing, Molly, it's *performance art.*" "It's a felony with a costume." "Exactly! Best kind." She stared at him for a long moment, weighing the absurdity of the plan against the very real fact that they were broke, hungry, and out of better options. Somewhere in the back of her mind, a small voice reminded her that stealing from children was a new low even for them. She silenced it with the memory of rent due Friday. "Fine," she said finally. "But I'm not wearing the cape." "You're wearing the cape." "I'm not wearing the cape, Bill." He was already draping it over her shoulders, humming a triumphant little tune, utterly unbothered by her protests, already halfway out the door toward a birthday party that had no idea what was coming for it.
created_at
"2026-07-01T07:44:46.675Z"
7/1/2026, 7:44:46 AM
model
"anthropic/claude-sonnet-5-20260630"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
7470
moderation_latency
(null)
generation_time
21344
tokens_prompt
77
tokens_completion
835
native_tokens_prompt
93
native_tokens_completion
1113
native_tokens_completion_images
(null)
native_tokens_reasoning
66
native_tokens_cached
0
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
num_fetches
(null)
web_search_engine
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"end_turn"
service_tier
"default"
usage
0.011316
router
(null)
provider_responses
0
endpoint_id
"37b16f45-eeae-45f3-9b85-2137f7c6d61b"
id
"msg_01DQHTJDVn39mExmBtz5xPWP"
is_byok
false
latency
3780
model_permaslug
"anthropic/claude-sonnet-5-20260630"
provider_name
"Amazon Bedrock"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.17.0; linux; x64))"
http_referer
(null)
request_id
"req-1782891886-KgoAnppyZqp74865bzdW"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1782891886-ByJ7JkJl25JhwpuKKWq7"
upstream_id
"msg_01DQHTJDVn39mExmBtz5xPWP"
total_cost
0.011316
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Amazon Bedrock"
response_cache_source_id
(null)
data_region
"global"
Evaluation details
Result
Evaluator
Details
Meta Data
97.63%
Matches word count
n/a
neededClean
false
words
493
0.00%
Dialogue to Total Word Ratio
Ratio: 51.90%, Deviation: 21.90%
neededClean
false
wordsTotal
499
wordsDialogue
259
48.8138%