NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCPWBXGQV8ASM39Q50B1CR6
Run Details
Model:
Claude Opus 5
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
21.2s
Cost:
$0.027715
Provider:
Amazon Bedrock
Evaluation summary
Score
Evaluator
Details Link
93.65%
Matches word count
(details)
41.84%
Dialogue to Total Word Ratio
(details)
67.7424%
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 napkin across the diner table and smoothed it flat with the side of her hand. "Here's the house," she said, drawing a lopsided square. "Here's the bouncy castle. Here's the presents." Bill leaned in. "Why's the bouncy castle bigger than the house?" "It's not to scale." "Nothing you draw is to scale. Last time you drew a bank and I thought it was a boat." "It was a boat, Bill. The money was on a boat." "Then why'd you draw a chimney on it?" Molly stabbed the napkin with her pen. "Focus. The party starts at two. Twenty-eight children, one clown, two hired parents—" "Hired parents?" "They're not the real parents, they're the ones who bring the extra folding chairs and pretend to like everybody. They leave by three." Bill sat back and folded his arms. "I want it on record that I have concerns." "You always have concerns." "I have concerns because we're robbing an eight-year-old." "We're robbing a seven-year-old. Her name's Priya. And she doesn't need forty presents, Bill. Forty. Her mother posted the list online. There's a telescope on it." "What's wrong with a telescope?" "Nothing's wrong with a telescope. That's the point. That telescope's worth four hundred quid and it's going to sit in a garage until 2041." Molly tapped the napkin twice. "We're not thieves. We're a correction." "We are thieves." "We're both." Bill picked up his coffee, then put it down without drinking it, which was how he signalled he was being serious. "How do we get in?" "That's the beautiful bit. Nobody checks. It's a children's party. You show up with a wrapped box and a tired expression and you're basically staff." She reached under the table and hauled up a shopping bag. "I got us two." "Two what?" "Two boxes. Wrapped. Empty." "They rattle." "They rattle because I put a Yorkie in each one so they'd feel legitimate. Don't eat them." "I've already eaten one." Molly closed her eyes. "When?" "In the car. It rattled and I got curious." Bill shrugged. "You'd have done the same." "I absolutely would have, that's not the issue." She flattened the napkin again. "Presents go on a trestle table by the patio doors. Two forty-five, the clown does the balloon animals. That's our window. Every eyeball in that garden will be pointed at a man making a giraffe." "And if he's rubbish at giraffes?" "Then we improvise." "You always say that." "Because it always happens." Molly slid a bin bag across the table, folded neat as a handkerchief. "You take the far end. Big stuff first, the telescope's in a box the size of a hedge. I'll take the envelopes." "There's envelopes?" "Bill. It's a birthday. There's always envelopes." Bill unfolded the bin bag, looked at it a long moment, and sighed. "One condition." "Go on." "We leave her one. The good one. The one she actually asked for." Molly considered this, then wrote *leave one* on the napkin in careful capitals, like a promise. "Fine," she said. "But not the telescope."
created_at
"2026-07-25T13:19:23.261Z"
7/25/2026, 1:19:23 PM
model
"anthropic/claude-opus-5-20260723"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1963
moderation_latency
(null)
generation_time
20893
tokens_prompt
75
tokens_completion
763
native_tokens_prompt
93
native_tokens_completion
1090
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"
service_tier
"default"
usage
0.027715
router
(null)
provider_responses
0
endpoint_id
"76cb4608-f48c-483d-8da8-9957fb44244e"
id
"msg_011CdNqqTpEYVT442o5GgpBs"
is_byok
false
latency
1961
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-1784985563-rrNxcsVse86gpgyfL02P"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784985563-EXuznYxhsUbQ86Ys6x9g"
upstream_id
"msg_011CdNqqTpEYVT442o5GgpBs"
total_cost
0.027715
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
93.65%
Matches word count
n/a
neededClean
false
words
509
41.84%
Dialogue to Total Word Ratio
Ratio: 74.17%, Deviation: 4.17%
neededClean
false
wordsTotal
515
wordsDialogue
382
67.7424%