NC Bench
Overview
About
Tests
Models
Vendors
Compare
Run: 2026-02-24-01KJ8VWHVTMK6KWMR24TTGR0HE
Run Details
Model:
Claude Opus 4.6 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
21.0s
Cost:
$0.020960
Provider:
Amazon Bedrock
Evaluation summary
Score
Evaluator
Details Link
99.38%
Matches word count
(details)
83.29%
Dialogue to Total Word Ratio
(details)
91.3330%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 70% 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 hand-drawn map across the hood of the van, smoothing out a wrinkle that bisected what appeared to be a bouncy castle. Bill leaned in, squinting. "Walk me through it again," he said. "And this time, skip the part where you compare yourself to Ocean's Eleven." "Fine. The party starts at two. The birthday boy is turning seven. His name is Connor, and according to my surveillance, he's a biter." "You surveilled a seven-year-old?" "I surveilled the perimeter, Bill. The kid just happened to be in it. Now, the presents will be stacked on a folding table near the back fence, right next to the piƱata station. We go in through the side gate at two-fifteen, when the sugar rush peaks and every adult in that backyard is questioning their life choices." Bill rubbed the bridge of his nose. "And nobody's going to notice two grown adults hauling a garbage bag full of presents through a children's party?" "That's the beauty of it. We won't look out of place because we'll be in disguise." Molly reached into the van and produced two matching purple polo shirts with the words PARTY PANDEMONIUM embroidered on the chest. "You're joking." "Party entertainment, Bill. We walk in like we own the place. If anyone asks, we're the backup magicians." "The backup magicians," he repeated flatly. "Every party has backup magicians." "No party in the history of civilization has ever had backup magicians." Molly waved him off. "You're overthinking this. I walk in, I distract the parents with balloon animals. You grab the presents, load them into the van. We're out in four minutes." "Can you even make balloon animals?" She paused just long enough for Bill to feel his stomach drop. "I can make a snake," she said. "A snake is just an uninflated balloon, Molly." "And it is a beautiful snake, Bill. The children love it." She jabbed a finger at the map. "Now focus. The real obstacle isn't the parents. It's the clown." Bill stared at her. "There's a clown?" "There's always a clown. His name is Mr. Giggles, and he is territorial. Last party I scoped, he chased a mime three blocks for stepping on his act. We need to neutralize him early." "I'm not fighting a clown." "You won't have to fight him. I've arranged for a colleague to call in a fake balloon emergency across town. Mr. Giggles will be gone by two-ten." "A fake balloon emergency," Bill said, his voice now barely a whisper. "Is that a thing?" "It is in the clown world. Those people take helium shortages very seriously." Bill looked at the map, then at the purple polo shirt, then at the cloudless sky above them as though searching for a merciful god. He pulled the shirt over his head. "If I get taken down by a seven-year-old biter, I'm naming you as an accomplice." Molly grinned and handed him a pair of oversized sunglasses. "That's the spirit."
created_at
"2026-02-24T22:21:32.42Z"
2/24/2026, 10:21:32 PM
model
"anthropic/claude-4.6-opus-20260205"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1386
moderation_latency
153
generation_time
20819
tokens_prompt
77
tokens_completion
816
native_tokens_prompt
92
native_tokens_completion
820
native_tokens_completion_images
(null)
native_tokens_reasoning
62
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
0.02096
router
(null)
provider_responses
0
endpoint_id
"a8708fb9-08ea-43bb-a3b1-165c87ef09df"
id
"d9a74ddf-9ac0-49ca-8344-463328355e51"
is_byok
false
latency
1347
model_permaslug
"anthropic/claude-4.6-opus-20260205"
provider_name
"Amazon Bedrock"
status
200
api_type
"completions"
id
"gen-1771971692-dym6U9o3Qj30nIymM4VH"
upstream_id
"d9a74ddf-9ac0-49ca-8344-463328355e51"
total_cost
0.02096
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Amazon Bedrock"
Evaluation details
Result
Evaluator
Details
Meta Data
99.38%
Matches word count
n/a
neededClean
false
words
495
83.29%
Dialogue to Total Word Ratio
Ratio: 71.91%, Deviation: 1.91%
neededClean
false
wordsTotal
502
wordsDialogue
361
91.3330%