NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-24-01KPZTY1PFJBDATMK8ESNBFV38
Run Details
Model:
Claude Opus 4.7
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
16.8s
Cost:
$0.028440
Provider:
Amazon Bedrock
Evaluation summary
Score
Evaluator
Details Link
14.30%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
7.1507%
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, pinning down the corners with ketchup bottles. "Okay, Bill. The target is 14 Maple Street. Sophie Pemberton's seventh birthday party. Saturday at two." Bill squinted. "Seven years old? Mol, that's dark." "It's a goldmine. Her dad's a hedge fund guy and the mom's an influencer. Do you know what kind of presents these kids get? Last year, one of them got a pony." "A pony?" "A pony, Bill." "How do you steal a pony?" "You don't steal the pony, you dingus. You steal the iPads, the LEGO sets, the gift cards. Rich-kid parties are lousy with gift cards." Bill rubbed his chin. "What's the theme?" "Does it matter?" "Of course it matters. If it's pirates, I wear an eyepatch. If it's princesses, I rent a tiara. You've gotta blend, Molly. That's Heisting 101." "It's unicorns." "Unicorns I can work with. I've got a horn in the trunk." "Why do you have a horn in the trunk?" "Molly, a man's trunk is his business." She rolled up the blueprints. "The bounce house goes up at noon. Clown arrives at one-thirty. Presents get piled on the sunroom table by two-fifteen. That's our window." "And the clown?" "What about him?" "Is he a threat?" "Bill. He's a clown." "Clowns can be ex-military, Molly. I knew a clown named Sergeant Bingo who could gut a man with a balloon animal." "Sergeant Bingo is not going to be there." "You don't know that." Molly sighed and signaled for more coffee. "Here's the plan. I go in as the face painter. I cased the actual face painter on Instagram, her name's Giselle, she's got a flower tattoo. I'll draw one on with eyeliner." "What do I do?" "You're the magician." Bill's face lit up. "I finally get to be the magician?" "Don't make it weird." "I'm bringing doves." "No doves." "Molly, a magician without doves is just a guy in a vest." "No doves, Bill! Doves poop everywhere and then we've got DNA evidence flapping around the sunroom." "Birds don't have DNA." "Birds absolutely have DNA!" The waitress refilled their cups and pretended not to listen. Bill leaned in. "Fine. No doves. But I'm doing the cup-and-ball trick. That's how we move the loot. Ball goes under the cup, cup goes in the bag, bag goes out the door." "That's... actually not terrible." "I have good ideas, Molly." "You had one good idea. Let's not gild the lily." "What about the kids? Seven-year-olds are vicious. They bite." "You've been doing this fifteen years and you're scared of second graders?" "I'm scared of anything under four feet tall. It's a documented phobia." Molly stood up and tossed a twenty on the table. "Saturday. Two o'clock. Unicorn horn, magician's vest, zero doves." "And if it goes sideways?" "Then we run, Bill. We run like the pony's chasing us."
created_at
"2026-04-24T13:30:12.06Z"
4/24/2026, 1:30:12 PM
model
"anthropic/claude-4.7-opus-20260416"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
632
moderation_latency
(null)
generation_time
16779
tokens_prompt
77
tokens_completion
761
native_tokens_prompt
98
native_tokens_completion
1118
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
0.02844
router
(null)
provider_responses
0
endpoint_id
"0fa65117-00d9-4835-808d-9b68041d9ce4"
id
"bf1ec862-1e5e-4ce7-8d53-f025a27b64a8"
is_byok
false
latency
632
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-1777037412-NaXw70FW6FSFUMYQ0ZRp"
session_id
(null)
api_type
"completions"
id
"gen-1777037412-QomWQSx2D4U2X61XHZyv"
upstream_id
"bf1ec862-1e5e-4ce7-8d53-f025a27b64a8"
total_cost
0.02844
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Amazon Bedrock"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
14.30%
Matches word count
n/a
neededClean
false
words
479
0.00%
Dialogue to Total Word Ratio
Ratio: 87.45%, Deviation: 17.45%
neededClean
false
wordsTotal
486
wordsDialogue
425
7.1507%