NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-05-31-01KSYXBHJVCZ51XHASKKV7E8Y2
Run Details
Model:
Claude Opus 4.8 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
9.9s
Cost:
$0.012390
Provider:
Anthropic
Evaluation summary
Score
Evaluator
Details Link
99.74%
Matches word count
(details)
52.20%
Dialogue to Total Word Ratio
(details)
75.9700%
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 wait until the piñata," Molly whispered. "Why the piñata?" "Because, Bill, when that thing bursts, every kid loses their tiny mind. Pure chaos. We grab the gifts and vanish." "What about the parents?" "Distracted. Filming. Crying about how fast their babies grew up." "And the clown?" "What clown?" "There's always a clown, Molly." "Then we neutralize the clown." "I'm not punching a clown." "You won't punch him. You'll simply offer to hold his balloons, and when his hands are full—" "I steal the balloons?" "No, you genius, we slip past him. Focus." "I don't like clowns. They've got those eyes." "Everyone's got eyes, Bill." "Not like clowns." "Can we please discuss the actual plan?" "Fine. How do we carry forty presents?" "The bouncy castle." "Excuse me?" "We deflate it, load the loot inside, roll it out like a giant burrito." "That's... actually brilliant." "I know." "One problem. I RSVP'd as 'Uncle Terrence.' What if they ask about Terrence?" "You said you were a guest?" "I brought a card. It plays music." "Bill." "It seemed festive!" "We are robbing children, not attending the party." "A man can do both, Molly. A man can do both."
created_at
"2026-05-31T11:40:22.505Z"
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
1032
moderation_latency
214
generation_time
9606
tokens_prompt
77
tokens_completion
342
native_tokens_prompt
93
native_tokens_completion
477
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
0.01239
router
(null)
provider_responses
0
endpoint_id
"b0d3beaa-5ff7-4666-a749-b1bb2b14c916"
id
"msg_01Q2f5SLvLXJ59LNcdppnK4s"
is_byok
false
latency
1032
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-tmwhyZrqEBUhvwi2TIgF"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1780227622-U5ktWuJei5TQpJs8ef71"
upstream_id
"msg_01Q2f5SLvLXJ59LNcdppnK4s"
total_cost
0.01239
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Anthropic"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
99.74%
Matches word count
n/a
neededClean
false
words
196
52.20%
Dialogue to Total Word Ratio
Ratio: 98.98%, Deviation: 8.98%
neededClean
false
wordsTotal
196
wordsDialogue
194
75.9700%