NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-02-08-01KGZ011FR2M4VJKWMH2JV0W3Q
Run Details
Model:
Claude Opus 4.5
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
9.1s
Cost:
$0.008400
Provider:
Anthropic
Evaluation summary
Score
Evaluator
Details Link
95.99%
Matches word count
(details)
96.73%
Dialogue to Total Word Ratio
(details)
96.3573%
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
Molly pressed her binoculars against the fence slats and whispered urgently. "Bill, we've got a problem. They hired a clown." "A clown? Nobody said anything about a clown." "I'm looking right at him. Big red nose. Floppy shoes. The whole nine yards." "This changes everything. Clowns are unpredictable, Molly. They're chaos agents." "I know what clowns are, Bill. I did three years in birthday party reconnaissance." "So what's the play? We abort?" "We don't abort. Those presents aren't going to steal themselves. Little Timmy's getting a PlayStation 5. I saw the wrapping paper." "How can you tell what's inside from the wrapping paper?" "Because I'm a professional, Bill. That's how." "Fine. What about the clown?" "You distract him. Ask for a balloon animal. Something complicated. A giraffe riding a motorcycle." "That's impossible." "Exactly. He'll be so focused on failing that I can slip through the back door and grab the goods." "And if he actually makes it?" "Then we recruit him. Anyone who can twist a giraffe riding a motorcycle is exactly the kind of talent we need." "I hate this plan." "You hate every plan." "And yet I keep showing up."
created_at
"2026-02-08T16:05:57.351059+00:00"
2/8/2026, 4:05:57 PM
model
"anthropic/claude-4.5-opus-20251124"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1577
moderation_latency
(null)
generation_time
8981
tokens_prompt
61
tokens_completion
264
native_tokens_prompt
75
native_tokens_completion
321
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)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
usage
0.0084
router
(null)
provider_responses
0
id
"msg_012y6FjyVSgvYMr64eHN2wPy"
status
200
is_byok
false
latency
1577
endpoint_id
"be883404-eb42-4b2d-b6e4-c7daa3aa8d62"
provider_name
"Anthropic"
model_permaslug
"anthropic/claude-4.5-opus-20251124"
api_type
"completions"
id
"gen-1770566747-PqEweuHKk3cq8SX9feiK"
upstream_id
"msg_012y6FjyVSgvYMr64eHN2wPy"
total_cost
0.0084
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Anthropic"
Evaluation details
Result
Evaluator
Details
Meta Data
95.99%
Matches word count
n/a
neededClean
false
words
192
96.73%
Dialogue to Total Word Ratio
Ratio: 94.27%, Deviation: 4.27%
neededClean
false
wordsTotal
192
wordsDialogue
181
96.3573%