NC Bench
Overview
About
Tests
Models
Vendors
Compare
Run: 2026-02-24-01KJ8TWR1229VBWE1EYYWRFEYN
Run Details
Model:
Claude Sonnet 4.6 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
9.5s
Cost:
$0.006666
Provider:
Amazon Bedrock
Evaluation summary
Score
Evaluator
Details Link
81.27%
Matches word count
(details)
98.31%
Dialogue to Total Word Ratio
(details)
89.7920%
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
"The target is a six-year-old's birthday party," Molly whispered, unrolling a hand-drawn map onto the park bench. "A what?" said Bill. "You heard me. Timmy Perkins. Turns seven on Saturday." "Six turning seven, Molly. That's our mark?" "The guest list alone has forty kids. Forty presents minimum. We're talking Lego sets, remote control cars, probably a bike." "We're stealing from a child." "We're redistributing gifts from a child who already has a trampoline, two iPads, and a golden retriever named Biscuit." "How do you know about Biscuit?" "I did reconnaissance." "You went to a child's birthday party for reconnaissance?" "I told them I was someone's aunt." "Whose aunt?" "I don't know, Bill, just a general aunt." "This is the worst plan you've ever had." "You said that about the art gallery." "We got chased by a swan at the art gallery." "And we still got the painting." Bill folded his arms. "What's our entry point?" "The bouncy castle. Nobody questions an adult climbing into a bouncy castle." "I question it." "Nobody normal, Bill." He stared at the map for a long moment. "Fine. But I'm keeping the Lego."
created_at
"2026-02-24T22:04:10.16Z"
2/24/2026, 10:04:10 PM
model
"anthropic/claude-4.6-sonnet-20260217"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
768
moderation_latency
292
generation_time
9156
tokens_prompt
77
tokens_completion
403
native_tokens_prompt
92
native_tokens_completion
426
native_tokens_completion_images
(null)
native_tokens_reasoning
85
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
0.006666
router
(null)
provider_responses
0
endpoint_id
"83d802cb-44f2-45a0-a9d5-cb9bf79777c8"
id
"706b53f5-69f4-4961-a326-0fd331a08f84"
is_byok
false
latency
765
model_permaslug
"anthropic/claude-4.6-sonnet-20260217"
provider_name
"Amazon Bedrock"
status
200
api_type
"completions"
id
"gen-1771970650-DcoytfVZvVuhEzIbsVZK"
upstream_id
"706b53f5-69f4-4961-a326-0fd331a08f84"
total_cost
0.006666
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Amazon Bedrock"
Evaluation details
Result
Evaluator
Details
Meta Data
81.27%
Matches word count
n/a
neededClean
false
words
188
98.31%
Dialogue to Total Word Ratio
Ratio: 86.39%, Deviation: 3.61%
neededClean
false
wordsTotal
191
wordsDialogue
165
89.7920%