NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCPT1MN6R88PQY6QJ7GN20F
Run Details
Model:
Claude Opus 5
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
21.1s
Cost:
$0.028115
Provider:
Amazon Bedrock
Evaluation summary
Score
Evaluator
Details Link
6.09%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
3.0466%
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 napkin across the table and smoothed it flat with two fingers. "Here's the house," she said, drawing a lopsided square in ketchup. "Here's the garden. Here's the bouncy castle." "Why does the bouncy castle matter?" "Everything matters, Bill. The bouncy castle is where the parents go to stop feeling old. It's a distraction hub. Ten adults minimum, all facing away from the patio doors." "You've been watching it." "Tuesday, Wednesday, and half of Thursday until the mum came out with a bin bag and I had to pretend to be a surveyor." Bill picked up a chip and pointed it at her. "Did she believe you?" "She asked about her boundary line. I said it was fine. She cried a bit." "So we go in through the patio doors." "We go in through the patio doors at fourteen hundred hours, which is when the entertainer starts. His name is Mister Wobbles and he is legally required to hold their attention for forty minutes." "Legally?" "Contractually. Same thing, tighter." Bill leaned back and folded his arms, which was his thinking posture, and also his indigestion posture. "How many presents?" "Thirty-one children invited. Assume twenty-eight attend, assume two bring nothing because their parents forgot, assume four bring vouchers, which we don't want because vouchers are traceable and frankly insulting." "So twenty-two boxes." "Twenty-two boxes, average value eleven pounds, and before you say it, I know." "I wasn't going to say anything." "You were going to say two hundred and forty-two pounds." "I was going to say two hundred and forty-two pounds is less than we made on the wedding." "The wedding had a cash bar, Bill. This is different. This is about the sideboard." He stopped chewing. "What sideboard?" "There's a sideboard in the dining room with a locked drawer, and on Wednesday the dad opened that drawer with a key he keeps behind the clock, and inside that drawer is a coin collection his father left him, and he sat looking at it for eleven minutes like a man at a graveside." "So the presents are—" "Cover. Camouflage. Nobody phones the police about a coin collection when there are twenty-two crying children on the lawn. Nobody even looks in the drawer for a week." Bill was quiet for a moment. "That's cruel, Moll." "It's a party bag economy. They'll all get replacements by Sunday, better ones, guilt-funded. The children will do better out of this than we will." "And the dad?" "The dad," Molly said, "keeps his father's coins behind a clock in a drawer he opens when nobody's watching. That's not owning something. That's haunting yourself." "That's the worst justification I've ever heard you make." "I know. I've been working on it since Thursday and it isn't landing." Bill sighed and pulled the napkin toward himself, turning it around, studying the ketchup. "Bin bags or the wheelie?" "Wheelie. Bin bags rustle and children have the hearing of bats." "And Mister Wobbles?" "Mister Wobbles," Molly said, standing and putting on her coat, "is going to have a puncture at half one, and we are going to be extremely helpful about it."
created_at
"2026-07-25T13:18:07.005Z"
7/25/2026, 1:18:07 PM
model
"anthropic/claude-opus-5-20260723"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1605
moderation_latency
(null)
generation_time
20971
tokens_prompt
75
tokens_completion
794
native_tokens_prompt
93
native_tokens_completion
1106
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
0.028115
router
(null)
provider_responses
0
endpoint_id
"76cb4608-f48c-483d-8da8-9957fb44244e"
id
"msg_011CdNqjqpth9EmWVYPgc9c4"
is_byok
false
latency
1604
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-1784985487-IBjnENgN9oZx443aMXfs"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784985487-swo7XxowM1BW19wXSejl"
upstream_id
"msg_011CdNqjqpth9EmWVYPgc9c4"
total_cost
0.028115
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
6.09%
Matches word count
n/a
neededClean
false
words
523
0.00%
Dialogue to Total Word Ratio
Ratio: 84.56%, Deviation: 14.56%
neededClean
false
wordsTotal
531
wordsDialogue
449
3.0466%