NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-24-01KQ0G2CEW4YBTEJMV8323MTPZ
Run Details
Model:
Claude Opus 4.7 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
14.0s
Cost:
$0.026140
Provider:
Amazon Bedrock
Evaluation summary
Score
Evaluator
Details Link
51.93%
Matches word count
(details)
97.08%
Dialogue to Total Word Ratio
(details)
74.5051%
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 back booth of Denny's smelled like burnt coffee and regret, which Molly felt was appropriate for the occasion. She unrolled a napkin across the table and smoothed it flat with the side of her hand, revealing a crayon diagram that looked like it had been drawn by someone mid-stroke. Bill squinted at it. He tilted his head. He tilted it the other way. "Is that the bouncy castle or a uterus?" "It's the bouncy castle, Bill." "Why's it got fallopian tubes?" "Those are the anchor ropes." Bill nodded slowly, the way a man nods when he has accepted that his life has brought him to this exact booth on this exact Tuesday. He reached for his coffee, found it empty, and sighed at the mug as though it had personally betrayed him. The party, Molly explained, was for a nine-year-old named Tanner whose father owned three car dealerships and a boat named *Liquid Assets*. The guest list was forty children, which meant forty gift bags, plus the pile on the present table, which Molly's reconnaissance suggested contained at least two drones, a Nintendo Switch, and something called a Hatchimal that retailed for ninety dollars and could apparently be flipped on eBay for double that before sundown. "Forty kids," Bill said. "Forty screaming, feral, sugar-loaded—" "Which is why we go in during pin-the-tail." "Pin-the-tail." "Blindfolds, Bill. Everyone's blindfolded. Even the parents get into it because there's sangria." Bill considered this. It was, he had to admit, elegant. He had been in this business for twenty-two years, and he had never once considered weaponizing a child's party game, which perhaps said more about his career than he was comfortable examining over an empty mug. "What's my cover?" "Clown." "No." "Bill." "Absolutely not." "You already own the shoes." "Those are orthopedic." Molly slid a photograph across the table. Bill stared down at a wig the color of a traffic cone and made a small, wounded sound. "I'll be the magician," she went on. "I do the disappearing act, except what disappears is the contents of the present table, into a false-bottom steamer trunk, which rolls out to the van during the piñata." "And the piñata is—" "Also full of presents. Smaller ones. Stocking stuffer tier." Bill rubbed his face. A waitress drifted past and refilled his coffee without asking, which he considered the single kindest thing anyone had done for him in months. "One question," he said. "What happens when the kid goes to open presents and there are no presents?" Molly smiled. She had a very specific smile for moments like this, a smile that had, over the years, preceded both their greatest triumphs and their worst felonies, often simultaneously. "We'll be in Ensenada," she said, "and Tanner will be building character." Bill picked up the napkin, folded it carefully, and tucked it into his jacket. "Fine," he said. "But I'm not honking a nose."
created_at
"2026-04-24T19:39:34.241Z"
4/24/2026, 7:39:34 PM
model
"anthropic/claude-4.7-opus-20260416"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
657
moderation_latency
(null)
generation_time
13976
tokens_prompt
77
tokens_completion
774
native_tokens_prompt
98
native_tokens_completion
1026
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
0.02614
router
(null)
provider_responses
0
endpoint_id
"0fa65117-00d9-4835-808d-9b68041d9ce4"
id
"e5f40066-fda0-4ab7-b820-99ca179e8e9d"
is_byok
false
latency
657
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-1777059574-5K1eSgRXgTnTwTxR5MZq"
session_id
(null)
api_type
"completions"
id
"gen-1777059574-1aElavOEL17BsfvsPpLB"
upstream_id
"e5f40066-fda0-4ab7-b820-99ca179e8e9d"
total_cost
0.02614
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Amazon Bedrock"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
51.93%
Matches word count
n/a
neededClean
false
words
484
97.08%
Dialogue to Total Word Ratio
Ratio: 30.77%, Deviation: 0.77%
neededClean
false
wordsTotal
494
wordsDialogue
152
74.5051%