NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-05-29-01KSSK5K2YXGXN7YNE94YFPEXH
Run Details
Model:
Claude Opus 4.8 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
20.9s
Cost:
$0.027940
Provider:
Anthropic
Evaluation summary
Score
Evaluator
Details Link
75.16%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
37.5778%
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 crayon-drawn map across the table, smoothing out the wrinkles with theatrical care. The paper was decorated with stickers of unicorns and a suspicious brown smudge. "This," she announced, "is the Hendersons' backyard. And this here is where they keep the loot." Bill squinted at the drawing. "Why does the treasure pile have a face?" "That's a piƱata, Bill. Try to keep up." He leaned back in his chair, arms crossed. "I just want it on record that I think robbing a seven-year-old's birthday party is a new low, even for us." "It's not robbing a seven-year-old," Molly corrected. "It's robbing approximately fifteen seven-year-olds. The presents, Bill. Think of the presents." "What's even in those presents? Building blocks? A doll that wets itself?" "Last month the Patterson kid got a tablet. A real one. And gift cards, Bill. Crisp, untraceable gift cards." She tapped the map. "Rich parents overcompensate. It's basic economics." Bill scratched his chin, clearly weakening. "Fine. How do we get past security?" "Security is one clown named Mister Bubbles and a mom with a juice box." "Don't underestimate Mister Bubbles," Bill said gravely. "Clowns are unpredictable." Molly ignored him and pointed to a stick figure in the corner. "I'll create a diversion. I'll dress as the magician they hired and 'accidentally' release the doves early." "They have doves?" "They will once I release them." Bill blinked. "Where are you getting doves?" "I have a guy." "You have a dove guy?" "I have many guys, Bill. That's why I'm in charge." She slid a second crayon drawing toward him. "While everyone's chasing birds, you slip in through the back, grab the gift table, and load it into the van." He studied the second picture. "Why am I dressed as a giant cupcake?" "Camouflage. There's a dessert theme." Bill set the paper down slowly. "Molly. I am a forty-three-year-old man. I have a criminal record in three states. And you want me to waddle into a child's party dressed as a baked good." "A delicious baked good," she said encouragingly. "Nobody suspects the cupcake." "Somebody always suspects the cupcake!" Molly leaned forward, eyes gleaming. "Look. We do this clean, we walk away with maybe two grand in toys and gadgets. We retire from the small stuff. No more vending machines, Bill. No more dignity-destroying parking meter raids." That landed. Bill's shoulders sagged with the weight of a thousand quarters. "The vending machine thing did break me a little," he admitted. "I know it did, sweetheart." He sighed, picking up the cupcake costume drawing again. "If a single child cries, I'm out." "They'll be too busy chasing doves to cry." "And if Mister Bubbles makes a move?" Molly smiled, sharp and confident. "Then Mister Bubbles learns why they call me the Birthday Bandit." Bill stared at her. "Nobody calls you that." "They will after Saturday." He groaned, but he was already reaching for the costume.
created_at
"2026-05-29T10:06:09.509Z"
5/29/2026, 10:06:09 AM
model
"anthropic/claude-4.8-opus-20260528"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
872
moderation_latency
(null)
generation_time
20902
tokens_prompt
77
tokens_completion
792
native_tokens_prompt
93
native_tokens_completion
1099
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
0.02794
router
(null)
provider_responses
0
endpoint_id
"dfc0e5bd-d703-4fe2-a7bb-655eb95d5441"
id
"msg_01EqbQhcMv7Y78hGoSTcPFqu"
is_byok
false
latency
872
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-1780049169-Z9yJRin0sF1CD6rdUNBz"
session_id
(null)
api_type
"completions"
id
"gen-1780049169-lp9PTDOmP5FQLb916dBu"
upstream_id
"msg_01EqbQhcMv7Y78hGoSTcPFqu"
total_cost
0.02794
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Anthropic"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
75.16%
Matches word count
n/a
neededClean
false
words
487
0.00%
Dialogue to Total Word Ratio
Ratio: 71.49%, Deviation: 21.49%
neededClean
false
wordsTotal
498
wordsDialogue
356
37.5778%