NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-05-31-01KSYXAJ4PQSEQB3KYSGYGYQH5
Run Details
Model:
Claude Opus 4.8 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
9.8s
Cost:
$0.013215
Provider:
Anthropic
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
51.59%
Dialogue to Total Word Ratio
(details)
75.7947%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 90% 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
"So we just walk in?" Molly asked. "Through the front door. Confidence is key." "Bill, it's a six-year-old's birthday party." "Exactly. Nobody suspects adults at a kid's party." "Everyone suspects adults at a kid's party. That's literally how parties work." "We'll bring a gift. Blend in." "We're stealing the gifts. Why would we bring one?" "Camouflage, Molly. You bring a gift, you take forty gifts. It's basic mathematics." "That's not mathematics. That's just stealing with extra steps." "What's the plan once we're inside?" "You distract them. Do a magic show." "I don't know any magic." "You pulled a wallet out of a stranger's pocket yesterday." "That's pickpocketing, not magic!" "To a six-year-old, what's the difference?" "And what'll you be doing during my so-called magic show?" "Loading presents into the bouncy castle." "You're going to steal the bouncy castle?" "It inflates, Bill. It's basically a getaway balloon." "That's the dumbest thing I've ever heard." "Coming from the man bringing a gift to a robbery." "Fine. What do we get the kid?" "Nothing expensive. We're taking it back anyway." "A re-gifted regret. How festive." "Just smile, wave, and follow my lead." "Molly?" "What?" "There's clowns. I hate clowns." "Then steal faster."
created_at
"2026-05-31T11:39:50.31Z"
5/31/2026, 11:39:50 AM
model
"anthropic/claude-4.8-opus-20260528"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
998
moderation_latency
(null)
generation_time
9720
tokens_prompt
77
tokens_completion
363
native_tokens_prompt
93
native_tokens_completion
510
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
0.013215
router
(null)
provider_responses
0
endpoint_id
"b0d3beaa-5ff7-4666-a749-b1bb2b14c916"
id
"msg_013ojSXNswbAvkFieJkTxiiB"
is_byok
false
latency
998
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-1780227590-5JVraITPHz3jzOiaBQDF"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1780227590-O28zPtlzCQ6LyG2fU2RM"
upstream_id
"msg_013ojSXNswbAvkFieJkTxiiB"
total_cost
0.013215
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Anthropic"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
199
51.59%
Dialogue to Total Word Ratio
Ratio: 99.02%, Deviation: 9.02%
neededClean
false
wordsTotal
204
wordsDialogue
202
75.7947%