NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCQQ6548KYGNHSPDF7FEBHC
Run Details
Model:
Claude Opus 5 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
23.9s
Cost:
$0.031165
Provider:
Amazon Bedrock
Evaluation summary
Score
Evaluator
Details Link
86.38%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
43.1902%
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
The blueprints were drawn on the back of a pizza box, which Bill considered his finest work to date. "Right," he said, tapping a grease stain with a pen. "This is the bouncy castle." Molly leaned in. "That's a pepperoni." "It's the bouncy castle *now*. Symbolically." "Symbolically." She pushed her sunglasses up into her hair, which was the closest she ever came to expressing despair. "Bill, we've knocked over three jewelers and a bank vault in Antwerp. We are professionals. And you've drawn a moat." "There *is* a moat. Sort of. There's an inflatable paddling pool and I'm not walking through it." "So we go around it." "That's what the moat *means*, Molly. That's what a moat is for." He circled a smear of cheese. "Present table. Two o'clock, north side of the garden, under the gazebo. Forty-two children have RSVP'd, which means forty-two presents minimum, and Tarquin's father is in hedge funds, so we're talking premium wrapping. Substantial gift bags. Possibly a scooter." "You want to steal a scooter from a seven-year-old." "I want to steal *forty-two* scooters from a seven-year-old. There's a difference and the difference is scale." Bill sat back, satisfied. "Now. Entry." "The side gate." "Locked." "Bill, it's a garden gate with a rope loop. My nan could open it and she's been dead since March." "The gate," Bill said, with dignity, "is covered by the sentry." Molly waited. "The what?" "Granny Pauline. Ninety-one. Sits in a wicker chair with a direct sightline to the gate, the gazebo, and the buffet. She hasn't moved from that chair in eleven years except to comment on people's weight." He produced a grainy photograph, taken from an alarming distance. "Eyes like a hawk. Hearing like a bat. She's the single greatest threat to this operation." "She's ninety-one." "So was Churchill." "Churchill wasn't ninety-one, and also he wasn't guarding a gazebo." "Details," said Bill. Molly took the photograph, studied it, and set it down. "Fine. What's the distraction?" "Magician." "You've hired a magician?" "I've hired *me*." Bill produced a top hat from beneath the table with a flourish that knocked over the milk. "Twenty minutes of close-up magic, all eyes on the patio, and while Granny Pauline is watching me produce a live dove from a child's ear—" "Where are you getting a live dove?" "—you load the presents into the wheelie bin and roll it out through the moat." "Through the moat." "You can go through the moat. You've got the shoes for it." Molly closed her eyes for a long moment. Somewhere in the flat, a kettle clicked off, which felt like a judgment. "Bill." "Mm." "They're presents. For children. Colouring books. Slime kits. Vouchers for a soft play centre in Croydon." "And a scooter." "And a scooter," she conceded. "Half each." "Half of a scooter." "Then we'll draw straws." Bill placed the top hat on his head, tilted it, and grinned at her with the terrible confidence of a man who had never once been arrested doing something sensible. "Party starts at two. Wear something with pockets."
created_at
"2026-07-25T13:34:01.902Z"
7/25/2026, 1:34:01 PM
model
"anthropic/claude-opus-5-20260723"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
4357
moderation_latency
(null)
generation_time
23797
tokens_prompt
75
tokens_completion
854
native_tokens_prompt
93
native_tokens_completion
1228
native_tokens_completion_images
(null)
native_tokens_reasoning
67
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
0.031165
router
(null)
provider_responses
0
endpoint_id
"76cb4608-f48c-483d-8da8-9957fb44244e"
id
"msg_011CdNrxJu2FiabmLjtavUVA"
is_byok
false
latency
2743
model_permaslug
"anthropic/claude-opus-5-20260723"
provider_name
"Amazon Bedrock"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.18.0; linux; x64))"
http_referer
(null)
request_id
"req-1784986441-4C5wBSkORrfipT0zEGbu"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784986441-PtiW6Wk4mv75TegClLQ3"
upstream_id
"msg_011CdNrxJu2FiabmLjtavUVA"
total_cost
0.031165
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
86.38%
Matches word count
n/a
neededClean
false
words
511
0.00%
Dialogue to Total Word Ratio
Ratio: 66.99%, Deviation: 16.99%
neededClean
false
wordsTotal
521
wordsDialogue
349
43.1902%