NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-24-01KPZV91WNF03K4RT7A1BQQZD1
Run Details
Model:
Claude Opus 4.7
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
16.9s
Cost:
$0.028540
Provider:
Amazon Bedrock
Evaluation summary
Score
Evaluator
Details Link
99.92%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
49.9595%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 50% 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 blueprint across the folding card table, weighing down the corners with half-eaten granola bars. The paper depicted, in meticulous crayon, the layout of 412 Maple Drive. "Okay, Bill. Walk me through it." Bill adjusted his ski mask, which he insisted on wearing during all planning sessions despite being indoors, in August, in his own apartment. "Right. So the target is Timothy Pemberton. Age: seven. Occasion: birthday party. Estimated present count: twenty-two, with a high-value probability of at least one LEGO set retailing north of ninety bucks." "How do you know all this?" "I'm on the Evite." Molly stared at him. "You're on the Evite." "I RSVP'd as 'Uncle Roger.' Plus one." "Bill, Timothy does not have an Uncle Roger." "Timothy has an Uncle Roger now, Molly, and Uncle Roger is bringing a guest, and Uncle Roger's guest is going to case the bouncy castle while Uncle Roger distracts the magician." Molly pinched the bridge of her nose. "There's a magician?" "Pockets Malone. Washed up. Does a dove trick but the dove died in 2019 so now it's just a sock." "Focus. How do we get the presents out?" Bill produced a laundry bag with a flourish. "Behold. Standard hamper, reinforced stitching, capable of holding approximately forty pounds of wrapped joy." "And nobody's going to notice two adults walking out with a sack of gifts?" "Which is why," Bill said, tapping the blueprint, "we strike at piñata o'clock. Peak chaos. Sugar-crazed children swinging a stick blindfolded at a papier-mâché unicorn. The parents will be filming. We enter the gift corner, we extract, we exit through the side gate where you'll be waiting with the minivan." "We have a minivan?" "I rented a minivan." "With what money?" "With the money we're about to make from the LEGO set." Molly sat down slowly. "Bill. I want to ask you something, and I want you to really think about the answer. Are we, perhaps, bad people?" Bill lifted his ski mask just enough to expose his mouth. "Molly. Listen. Timothy Pemberton's father is a hedge fund guy. Timothy's last birthday, he got a pony. An actual pony. It lives at a farm now because their backyard wasn't zoned for equine. We are not villains. We are economic rebalancers." "We're stealing from a child." "We're stealing from a child who has a *pony.*" Molly considered this. "There's going to be a Nintendo Switch in there, isn't there." "Evite said 'no gifts over one hundred dollars,' which in rich people means minimum two hundred. There will be a Switch. There will possibly be two Switches." "Fine." She rolled up the blueprint. "But I'm not hurting the magician." "Agreed. Pockets has been through enough." "And if things go sideways—" "We grab what we can and retreat through the bouncy castle. Soft landings, minimal footprints, the children will think we're part of the entertainment." Molly extended her hand. Bill shook it solemnly, his ski mask already rolling back down over his face.
created_at
"2026-04-24T13:36:12.707Z"
4/24/2026, 1:36:12 PM
model
"anthropic/claude-4.7-opus-20260416"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
630
moderation_latency
(null)
generation_time
16790
tokens_prompt
77
tokens_completion
793
native_tokens_prompt
98
native_tokens_completion
1122
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
0.02854
router
(null)
provider_responses
0
endpoint_id
"0fa65117-00d9-4835-808d-9b68041d9ce4"
id
"e980afae-bd0e-43c1-8a67-79a12dc2f7e1"
is_byok
false
latency
630
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-1777037772-eVgbR4OFoyUozq3oCeVI"
session_id
(null)
api_type
"completions"
id
"gen-1777037772-xM05nqRbbKROR7Wdk7OD"
upstream_id
"e980afae-bd0e-43c1-8a67-79a12dc2f7e1"
total_cost
0.02854
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Amazon Bedrock"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
99.92%
Matches word count
n/a
neededClean
false
words
497
0.00%
Dialogue to Total Word Ratio
Ratio: 76.49%, Deviation: 26.49%
neededClean
false
wordsTotal
502
wordsDialogue
384
49.9595%