NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-01-01KWE9Y3XPGRDH8GZ1HBZVB1DV
Run Details
Model:
Claude Sonnet 5
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
17.2s
Cost:
$0.010176
Provider:
Amazon Bedrock
Evaluation summary
Score
Evaluator
Details Link
99.92%
Matches word count
(details)
0.02%
Dialogue to Total Word Ratio
(details)
49.9716%
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
The back of the van smelled like stale coffee and regret, which Bill supposed was appropriate given their line of work. "So walk me through it again," Molly said, tapping the folded invitation against her knee. "It's a six-year-old's birthday party, Molly. There is nothing to walk through. We go in, we find the present table, we leave." "Nothing to walk through, he says. Bill. Bill. There will be children." "I'm aware there will be children." "Small, loud, extremely observant children. Have you ever tried to lie to a six-year-old? They see through everything. They have no social contract yet. No filter. One of them is gonna point right at you and say 'that man is stealing our presents' and then what?" "Then I say, 'No, sweetie, I'm the entertainment,' and I do a little shuffle." "You don't know how to shuffle." "I know how to shuffle." "You have the rhythm of a filing cabinet." Bill exhaled through his nose and adjusted the bow tie he'd stolen from a dry cleaner two towns over. "The plan is solid. I go in dressed as the magician they hired—" "Who you also hired to not show up." "Who I bribed to not show up, yes, and you come in twenty minutes later as the caterer's assistant, and while everyone's watching me pull a scarf out of my sleeve, you load the presents into the laundry cart and wheel it out to the side door." "And what if the real caterer's assistant is there?" "There is no real caterer's assistant. I checked. It's just the one guy, and he's allergic to literally everything, including apparently nuts, dairy, and I think glitter, so he called in sick this morning after eating the cupcake I left on his porch." Molly stared at him. "You poisoned a caterer." "I inconvenienced a caterer. There's a difference." "There is not a difference, Bill." "He'll be fine. He's got an EpiPen. Probably." She rubbed her temples like she could massage the plan into something that made sense. "Okay. Fine. Say we get the presents. Say the shuffle works, say the kids don't smell a rat, say the allergic man lives. What exactly are we stealing? It's a six-year-old's birthday party. It's gonna be Legos and dinosaur pajamas and maybe, if we're very lucky, a gift card to a toy store." "The father is Victor Alcante." Molly went quiet. "The Victor Alcante," Bill said, letting it sit there. "The one who owes half the East Side and skipped town for eight months?" "That Victor Alcante, who apparently loves his kid so much he flew back into the country under a fake name just to watch him blow out six candles, and who, according to my guy, is currently sitting on a duffel bag full of cash disguised as a very large, very heavy present labeled 'From Dad, Do Not Open Yet.'" Molly blinked. Then smiled, slow and dangerous. "Well," she said, "in that case. Where's my apron?"
created_at
"2026-07-01T07:40:14.145Z"
7/1/2026, 7:40:14 AM
model
"anthropic/claude-sonnet-5-20260630"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1995
moderation_latency
(null)
generation_time
17074
tokens_prompt
77
tokens_completion
743
native_tokens_prompt
93
native_tokens_completion
999
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
0.010176
router
(null)
provider_responses
0
endpoint_id
"37b16f45-eeae-45f3-9b85-2137f7c6d61b"
id
"msg_015gY5KgiCbdVPPtMHSuqkfx"
is_byok
false
latency
1995
model_permaslug
"anthropic/claude-sonnet-5-20260630"
provider_name
"Amazon Bedrock"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.17.0; linux; x64))"
http_referer
(null)
request_id
"req-1782891614-As1XLuUDAL8trEm5W5zL"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1782891614-Kg0wYevdBlWeQZC5qfPK"
upstream_id
"msg_015gY5KgiCbdVPPtMHSuqkfx"
total_cost
0.010176
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Amazon Bedrock"
response_cache_source_id
(null)
data_region
"global"
Evaluation details
Result
Evaluator
Details
Meta Data
99.92%
Matches word count
n/a
neededClean
false
words
497
0.02%
Dialogue to Total Word Ratio
Ratio: 82.90%, Deviation: 12.90%
neededClean
false
wordsTotal
503
wordsDialogue
417
49.9716%