NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-05-31-01KSYXBHJ6SHP87WBBYDN3RNPD
Run Details
Model:
Claude Opus 4.8 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
19.5s
Cost:
$0.028365
Provider:
Anthropic
Evaluation summary
Score
Evaluator
Details Link
97.63%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
48.8138%
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 kitchen table, smoothing the wrinkles with the flat of her hand. "Behold," she said. "The Hendersons' backyard. Site of tomorrow's score." Bill leaned over, squinting. "This is drawn on a napkin." "It's a very detailed napkin." "There's a coffee stain right where you've marked the entrance." "That's not a stain, that's the bouncy castle." Molly tapped it proudly. "Our point of ingress. We come over the back fence at exactly two o'clock, while the kids are distracted by the magician." Bill folded his arms. "And what makes you so sure there's a magician?" "Because it's Timmy Henderson's eighth birthday, and I read the flyer his mother posted in the grocery store. Balloon animals, face painting, and a magician named the Great Zambino." "You did reconnaissance at a grocery store?" "I did reconnaissance everywhere, Bill. That's what separates professionals from amateurs." He picked up the napkin and held it to the light, as if expecting hidden ink. "Let me make sure I understand the plan. We, two grown adults, are going to crash a child's birthday party to steal a pile of presents." "Not a pile. The pile. Do you know what kids get these days? Tablets. Gaming consoles. Gift cards stuffed inside little envelopes shaped like dinosaurs." "And the parents?" "Will be too busy filming everything for social media to notice two caterers slipping out the side gate with a duffel bag." Bill set the napkin down slowly. "We're caterers now." "We're whatever the situation demands." She slid a paper plate across the table. "I made us disposable name tags. You're 'Chad, Catering Associate.' I'm 'Brenda, Team Lead.'" "Why do you get to be Team Lead?" "Because I drew the napkin." Bill rubbed his temples. "Molly, the last time we improvised a disguise, you ended up handcuffed to a vending machine." "That was different. That was a bank. This is a backyard full of seven-year-olds high on sheet cake. Where's the danger?" "The danger," Bill said, "is that one of those seven-year-olds bites my ankle and I get tetanus." "Wear thick socks." He stared at her for a long, defeated moment. "And the getaway?" "My nephew's ice cream truck. Idling out front. Nobody questions an ice cream truck near a kids' party. It's practically camouflage." "Your nephew agreed to this?" "My nephew thinks we're delivering ice cream." Bill exhaled through his nose, the sound of a man watching his life choices parade past him. "So to summarize. We dress as caterers, infiltrate a magic show, steal a duffel bag of dinosaur envelopes, and escape in a stolen ice cream truck driven by an unwitting relative." "Borrowed ice cream truck," Molly corrected. "And yes. Flawless, isn't it?" "It is the single worst plan I have ever heard." "But?" Bill picked up the name tag, peeled the backing, and stuck it crookedly to his chest. "But I haven't got rent." Molly grinned. "Welcome aboard, Chad." "I hate you, Brenda."
created_at
"2026-05-31T11:40:22.484Z"
5/31/2026, 11:40:22 AM
model
"anthropic/claude-4.8-opus-20260528"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
853
moderation_latency
(null)
generation_time
19417
tokens_prompt
77
tokens_completion
789
native_tokens_prompt
93
native_tokens_completion
1116
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
0.028365
router
(null)
provider_responses
0
endpoint_id
"b0d3beaa-5ff7-4666-a749-b1bb2b14c916"
id
"msg_013nXk6rpH3xBThBANt3A7ii"
is_byok
false
latency
853
model_permaslug
"anthropic/claude-4.8-opus-20260528"
provider_name
"Anthropic"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.16.0; linux; x64))"
http_referer
(null)
request_id
"req-1780227622-17sWJU1I7s6WqdtDFd3M"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1780227622-Tmg52i5GkyrlaRMh8hAx"
upstream_id
"msg_013nXk6rpH3xBThBANt3A7ii"
total_cost
0.028365
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Anthropic"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
97.63%
Matches word count
n/a
neededClean
false
words
493
0.00%
Dialogue to Total Word Ratio
Ratio: 77.26%, Deviation: 27.26%
neededClean
false
wordsTotal
497
wordsDialogue
384
48.8138%