NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCQRX25EE0J6B49FJR8CA7A
Run Details
Model:
Claude Opus 5 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
21.1s
Cost:
$0.029165
Provider:
Amazon Bedrock
Evaluation summary
Score
Evaluator
Details Link
35.00%
Matches word count
(details)
0.02%
Dialogue to Total Word Ratio
(details)
17.5089%
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 blueprint across the hood of the van and smoothed it flat with the side of her hand. "That's a napkin," Bill said. "It's a *schematic*." "It's a napkin from a Denny's, Molly. It's got syrup on it." "That's the moat." Bill leaned in. "There's a moat?" "There's a bouncy castle, and bouncy castles have moats, because the parents don't want the kids running off, so they ring the thing with those little plastic fences. Which, for our purposes, is a moat." "For our purposes." Bill straightened up and rubbed his eyes with both fists, the way he did when he was trying not to say something. "Molly. Sweetheart. Light of my criminal life. We used to do banks." "We did *one* bank." "We did one bank, and it was beautiful, and now we're stealing a Nintendo Switch off a seven-year-old named Braxton." "Braxton's turning eight, and he's got a hundred and forty guests, and each of those guests brings a present averaging thirty-two dollars retail. That's four thousand four hundred and eighty dollars of merchandise sitting on a folding table under a banner, guarded by one clown." "A clown." "Gerald. Party clown, twenty years' experience, bad knee." Molly tapped the napkin. "Gerald goes on at two-fifteen for balloon animals. That's our window. Every eyeball in that backyard swings to Gerald, and the present table becomes, functionally, unattended." "And the parents?" "Drinking. There's a mimosa situation on the patio. Mothers of eight-year-olds at a two o'clock party, Bill, they will not be operating at peak vigilance." Bill was quiet for a second. "How do we move four grand of Peppa Pig out of a suburb?" "Piñata." "Say that again." "The piñata comes in on a dolly. Big one, five feet, shaped like a dragon. I called the rental place pretending to be the caterer. It arrives at one-fifty, and it arrives *empty*, because they fill it on site." "You want to load presents into the piñata." "I want to load presents into the piñata, wheel the piñata out to the van as though we are removing the piñata for reasons of piñata safety, and drive away while a hundred and forty children watch a clown make a wiener dog." Bill looked at the napkin for a long time. "That's actually pretty good," he admitted. "Thank you." "That's the best thing you've ever said." "I know." "What are we wearing?" "Aprons. Nobody stops an apron at a children's party. An apron is a badge." Molly folded the napkin and slid it into her jacket. "You're on the dolly. I'm on the table. If anybody asks, we're with the bakery." "And if Gerald asks?" "Gerald won't ask. Gerald is a professional and he has forty minutes of material to get through with a bad knee." "And if a kid asks?" Molly opened the driver's door and paused. "Kids don't ask, Bill. Kids *tell*. Which is why, if a child engages you, you look them dead in the eye and you say, 'There's cake inside.'" "Is there cake inside?" "There's always cake inside." She got in. "Get the aprons."
created_at
"2026-07-25T13:34:58.127Z"
7/25/2026, 1:34:58 PM
model
"anthropic/claude-opus-5-20260723"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
3124
moderation_latency
(null)
generation_time
20939
tokens_prompt
75
tokens_completion
819
native_tokens_prompt
93
native_tokens_completion
1148
native_tokens_completion_images
(null)
native_tokens_reasoning
54
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
0.029165
router
(null)
provider_responses
0
endpoint_id
"76cb4608-f48c-483d-8da8-9957fb44244e"
id
"msg_011CdNs2NkobMtKWesL5NUxw"
is_byok
false
latency
1381
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-1784986498-cED9kadcFevWZAUfFU1O"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784986498-9u9cSrmZU87gGkOj98hZ"
upstream_id
"msg_011CdNs2NkobMtKWesL5NUxw"
total_cost
0.029165
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
35.00%
Matches word count
n/a
neededClean
false
words
518
0.02%
Dialogue to Total Word Ratio
Ratio: 83.24%, Deviation: 13.24%
neededClean
false
wordsTotal
525
wordsDialogue
437
17.5089%