NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCPTA0DAQD5R5CG8E72HHTH
Run Details
Model:
Claude Opus 5
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
21.4s
Cost:
$0.027640
Provider:
Amazon Bedrock
Evaluation summary
Score
Evaluator
Details Link
95.99%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
47.9943%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 50% 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 van smelled like old vinyl and older nachos. Molly had the blueprints spread across the dashboard, though calling them blueprints was generous — she'd traced the layout of the Pemberton backyard onto the back of a placemat from a pancake house. "Right," she said. "Bounce house here. Piñata here. Presents on the gift table, northwest corner, under the awning." Bill squinted at the placemat. "Why's there syrup on the bounce house?" "That's not syrup, that's the perimeter fence." "It's shiny." "It's a fence, Bill." He leaned back and folded his arms over a stomach that had defeated three prison diets. "I want it noted that I raised concerns." "Noted." "Formally noted." "There's no form, Bill. There's no committee. It's you, me, and forty-two seven-year-olds." "Forty-two." He whistled low. "That's a lot of little witnesses." "Children are unreliable witnesses. It's documented. One of them will tell the police the thief was a dinosaur." "That's if they even notice. Kids at a party, they're feral. Sugar-blind." He pulled a clown nose out of his jacket pocket and turned it over, mournful. "I still say we go in as entertainment." Molly stared at the nose. "You are not going in as a clown." "Balloon animals get you access, Molly. Balloon animals get you *trust*." "You can't make balloon animals." "I can make a snake." "A snake is a balloon." "It's a snake if I say it's a snake." He put the nose on. It sat crookedly. "Look at me. Would you suspect this face?" "I'd suspect that face of everything ever done in this country." "That's the beauty of it." She took the nose off him and shoved it in the glovebox, where it joined a flashlight, a bag of zip ties, and a birthday card they'd bought for four dollars as cover. "Here's the plan. Two-fifteen, the magician arrives. Everyone goes inside for the show. That's our window — six minutes, maybe eight if the magician's any good." "And if he's bad?" "Then we've got four minutes and a yard full of disappointed children, which is worse than dogs." Bill considered this. "What about the mother? The one with the clipboard." "Deborah." "You know her name?" "I've been watching this house for nine days, Bill, of course I know her name. Deborah runs the party like a military operation. She's got a laminated schedule. She's got a *timer*." "I respect her." "So do I. That's why we're not going anywhere near her." Molly folded the placemat and tucked it into her coat. "You take the gate. I take the table. Anything wrapped, it goes in the bag. Don't get sentimental about it." "When have I ever gotten sentimental." "The bicycle." "That was a *nice bicycle*." "You cried, Bill." "It had a bell." He looked out the window at the distant flutter of streamers, the sad sag of a half-inflated bounce house, the woman with the clipboard pointing at a folding chair like it had insulted her. "Two-fifteen?" "Two-fifteen." "Fine." He reached back into the glovebox. "But the nose comes with me."
created_at
"2026-07-25T13:18:15.572Z"
7/25/2026, 1:18:15 PM
model
"anthropic/claude-opus-5-20260723"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2054
moderation_latency
(null)
generation_time
21338
tokens_prompt
75
tokens_completion
768
native_tokens_prompt
93
native_tokens_completion
1087
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
0.02764
router
(null)
provider_responses
0
endpoint_id
"76cb4608-f48c-483d-8da8-9957fb44244e"
id
"msg_011CdNqkUWmGnMSrwR48WW2C"
is_byok
false
latency
2054
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-1784985495-obzVDm2hWZ6MMCPnUmMZ"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784985495-wESmgL5G4QBi1QpdPWYf"
upstream_id
"msg_011CdNqkUWmGnMSrwR48WW2C"
total_cost
0.02764
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
95.99%
Matches word count
n/a
neededClean
false
words
508
0.00%
Dialogue to Total Word Ratio
Ratio: 64.80%, Deviation: 14.80%
neededClean
false
wordsTotal
517
wordsDialogue
335
47.9943%