NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-24-01KPZVA2JBGVQECD2FXMS9VX8Z
Run Details
Model:
Claude Opus 4.7
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
16.5s
Cost:
$0.029040
Provider:
Amazon Bedrock
Evaluation summary
Score
Evaluator
Details Link
99.98%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
49.9928%
Overall
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 spread the blueprints across the sticky diner table, shoving aside a plate of half-eaten pancakes. Bill leaned in, squinting like a man who needed glasses but was too cheap to admit it. "Okay, genius," Molly said. "Walk me through it one more time." "We wait until the piñata." "Why the piñata?" "Because, Molly, the piñata is the moment of maximum chaos. Kids screaming, candy flying, some dad trying to look cool with a Wiffle ball bat. Nobody's watching the gift table." "And then?" "And then I roll in with the pillowcase." "You roll in." "Tactically." "Bill, you're forty-six." "I'm forty-four, and I can still roll." Molly took a long sip of coffee. "Last time you 'rolled,' you threw out your shoulder at a Chuck E. Cheese." "That was a skee-ball injury and you know it." "Uh huh." Bill tapped the blueprint. "The gift table is here, by the bounce house. Two exits. One through the kitchen, one through the backyard gate. The mom—" "Deborah." "Deborah. She'll be distracted refilling juice boxes around three-fifteen." "How do you know that?" "I read her Pinterest." Molly stared at him. "You read her Pinterest." "She has a whole board, Molly. 'Liam's Dino-Mite 6th Birthday.' Every snack has a toothpick with a little flag. The woman is a military operation." "And we're stealing from her." "We're stealing from a six-year-old named Liam, technically." "Don't say it like that." "You said it like that first!" Molly rubbed her temples. "Okay. Fine. I go in as the face painter." "You don't know how to paint faces." "I painted your face on Halloween." "You made me look like a sad raccoon." "That was the costume, Bill, you were a sad raccoon." "You could've told me that before the party!" Molly held up a hand. "Focus. I distract the kids. You get the presents. We rendezvous at the van. We're out before the cake hits the table." "What about the clown?" "What clown?" "Deborah hired a clown. Giggles McMurphy." "Giggles McMurphy is not a real name." "Tell that to his DBA filing, which I looked up, because I am a professional." "Is he a problem?" Bill hesitated. "He might be ex-military." "He might be *what*?" "There's a forum. Look, it doesn't matter. We stay away from Giggles, we grab the loot, we leave. The presents alone gotta be worth four grand. That one kid's parents are lawyers. Lawyers buy the big LEGO sets. The ones with the numbers." "The numbers?" "The set numbers, Molly. The four-digit ones. Those are the good ones." Molly folded up the blueprints. "I can't believe my life has come to this." "To what?" "Robbing a kindergartener named Liam while a possibly-Special-Forces clown named Giggles watches." Bill grinned and dropped a crumpled ten on the table. "Babe, that's not the low point. The low point is we need the money because we lost the last score betting on a chihuahua race." "Don't remind me." "Paco deserved to win." "Paco fell asleep, Bill."
created_at
"2026-04-24T13:36:46.168Z"
4/24/2026, 1:36:46 PM
model
"anthropic/claude-4.7-opus-20260416"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
740
moderation_latency
(null)
generation_time
16381
tokens_prompt
77
tokens_completion
813
native_tokens_prompt
98
native_tokens_completion
1142
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)
num_fetches
(null)
web_search_engine
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"end_turn"
usage
0.02904
router
(null)
provider_responses
0
endpoint_id
"0fa65117-00d9-4835-808d-9b68041d9ce4"
id
"21e1904b-6a5a-4d8f-90e3-9e982f8a9759"
is_byok
false
latency
740
model_permaslug
"anthropic/claude-4.7-opus-20260416"
provider_name
"Amazon Bedrock"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.1; linux; x64))"
http_referer
(null)
request_id
"req-1777037806-G1wMbVGO5VJjIMVTL5jo"
session_id
(null)
api_type
"completions"
id
"gen-1777037806-hwdI2XSl1OEOoum3TYum"
upstream_id
"21e1904b-6a5a-4d8f-90e3-9e982f8a9759"
total_cost
0.02904
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Amazon Bedrock"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
99.98%
Matches word count
n/a
neededClean
false
words
498
0.00%
Dialogue to Total Word Ratio
Ratio: 84.87%, Deviation: 14.87%
neededClean
false
wordsTotal
509
wordsDialogue
432
49.9928%