NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCQX6KR3N86MKSER6JRTA35
Run Details
Model:
Claude Opus 5 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
24.6s
Cost:
$0.031190
Provider:
Amazon Bedrock
Evaluation summary
Score
Evaluator
Details Link
27.17%
Matches word count
(details)
0.01%
Dialogue to Total Word Ratio
(details)
13.5868%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 30% 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 blueprints were drawn on the back of a pizza box, which Bill considered the height of professionalism, since the last job had been planned on a napkin that dissolved in the rain. Molly leaned over the table, tapping the greasy cardboard with a chewed pencil. The kitchen light swung above them, casting shadows that made the whole enterprise feel more sinister than it deserved. Somewhere below, a neighbor's television murmured about weather. "Bounce house here," she said. "Cake table here. Presents stacked on the piano bench, west wall, under the balloon arch." Bill squinted. "Why's the piano bench got a skull drawn on it?" "That's a strategic marker." "It looks like a skull." "It's a strategic marker with charisma." The target was the seventh birthday of one Timothy Pemberton-Frost, whose father imported marble and whose mother had sent out invitations printed on cardstock thick enough to stop a bullet. Molly had learned all this from a leaflet at the stationery shop, which she considered reconnaissance and Bill considered nosiness. Forty children. Forty gifts. The Pemberton-Frosts' social circle did not do gift bags; they did *tokens of esteem*, which in practice meant tablets, remote-control helicopters, and one rumored pony deposit certificate. Bill scratched the back of his neck. "I want it on the record that stealing from children makes me feel unwell." "You stole a wheelchair last April." "It was unoccupied and it was going downhill anyway." Molly slid a photograph across the table: the Pemberton-Frost side gate, wrought iron, unlocked in every one of her four surveillance passes, because rich people believed crime happened elsewhere, like Belgium. Beside the gate, a catering entrance. Beside the catering entrance, a stack of collapsed cardboard boxes, which was practically an engraved invitation. "Costumes," she said, and produced two bundles from a bin bag. Bill unfolded his. He held it at arm's length, the way one holds a dead thing found behind a fridge. It was a full-body foam suit, magenta, with a beak. "What is it." "A flamingo." "Why is it a flamingo?" "Entertainment arrives at two-fifteen. Children's entertainer, an act called Fabulous Flamingo Frank. Frank had an accident on his moped this morning." Bill lowered the costume. "Did Frank have an accident, Molly, or did Frank *have an accident*." "Frank had a genuine accident and I sent flowers." She unwrapped her own costume, which was a second flamingo, slightly larger and missing one eye. "We do twenty minutes of balloon animals, we suggest a game of hide-and-seek, forty children scatter, nobody's watching the piano bench, and we load the boxes into the catering van." "We don't have a catering van." "We have a catering van at one-forty." Bill sat down heavily. He was thinking about his mother, who had wanted him to be a dentist, and about how far a person could drift from the original blueprints of a life. "Can I be the one who does the voice?" "Which voice?" "The flamingo voice. I've been working on it." Molly considered. Outside, rain began, apologetically. "Go on then," she said. "But keep it under the balloon noise. Frank's a professional."
created_at
"2026-07-25T13:37:18.98Z"
7/25/2026, 1:37:18 PM
model
"anthropic/claude-opus-5-20260723"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
3884
moderation_latency
146
generation_time
24386
tokens_prompt
75
tokens_completion
863
native_tokens_prompt
93
native_tokens_completion
1229
native_tokens_completion_images
(null)
native_tokens_reasoning
50
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
0.03119
router
(null)
provider_responses
0
endpoint_id
"76cb4608-f48c-483d-8da8-9957fb44244e"
id
"msg_011CdNsCne2F6GwZuDVdNGFD"
is_byok
false
latency
1770
model_permaslug
"anthropic/claude-opus-5-20260723"
provider_name
"Amazon Bedrock"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.18.0; linux; x64))"
http_referer
(null)
request_id
"req-1784986638-alUfc2t9TZSYteHfXnbt"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784986638-nFOikepb5nhlLUkaBZTk"
upstream_id
"msg_011CdNsCne2F6GwZuDVdNGFD"
total_cost
0.03119
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
27.17%
Matches word count
n/a
neededClean
false
words
519
0.01%
Dialogue to Total Word Ratio
Ratio: 43.75%, Deviation: 13.75%
neededClean
false
wordsTotal
528
wordsDialogue
231
13.5868%