NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCR9A8DHANYVS0Z28C8CYWS
Run Details
Model:
Claude Opus 5 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
23.7s
Cost:
$0.029815
Provider:
Amazon Bedrock
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
49.9996%
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 blueprints were drawn on the back of a pizza box, which Molly considered a step up from their last job, planned entirely on a napkin that Bill later used to blow his nose. "Right," she said, tapping the greasy cardboard with a pen. "Fourteen hundred hours. Bouncy castle goes up here, by the hedge. Cake table's here. And the presents—" she circled a smear of tomato sauce, "—stack up on the patio, under the gazebo." Bill leaned in, chewing the last crust. "How many presents we talking?" "It's Tarquin Bletchley-Frome's seventh birthday. His father owns three car dealerships and a resentment toward his own childhood. There will be presents, Bill. There will be an *avalanche* of presents." "And the security?" "Twenty-two children under the age of eight, one clown, and a woman named Fiona who has been drinking prosecco since eleven." Bill sat back, deeply satisfied. "So no security." "I didn't say that." Molly's pen hovered. "There's a magician. Marvo the Magnificent. He does the two till two-forty slot." "So?" "So he's ex-Special Forces." Bill choked on his crust. "He's *what*?" "Turned to close-up magic after a difficult tour. Very sad story, actually, I read about it in the parish newsletter. Point is, we go in at two-fifteen, when he's mid-routine and both his hands are inside a top hat." "Why don't we just wait till he's gone?" "Because at two-forty they open the presents." Molly said it slowly, the way you'd explain gravity to a pigeon. "Once they're open, they're not presents anymore, Bill. They're just *stuff*. Stuff belonging to a child. That's theft." "And this isn't?" "This is *interception*." Bill considered this and found the logic airtight, which was how most of their convictions had begun. He pulled a paper bag from his jacket and produced two red foam noses. "Costumes," he announced. "I got us clown noses." "There's already a clown." "There can be two clowns." "There cannot be two clowns, Bill, that's how you get a *situation*. Children can count." Molly pinched the bridge of her nose. "We go as caterers. White shirts, black trousers, big flat trays. Nobody looks at a caterer. Nobody's ever looked at a caterer in the history of parties." "I look at caterers." "You look at sausage rolls, which happen to be *near* caterers." Bill turned the nose over in his fingers, wounded. "Fiver each, these." "Keep the receipt." She stabbed the pizza box again. "Present pile onto the trays, trays into the wheelie bin, wheelie bin down the side alley to the van. Ninety seconds, in and out, and we're eating chips in Margate by four." "And if someone stops us?" "Then you smile, you say 'gluten-free option,' and you keep walking. Nobody argues with gluten." Bill nodded slowly, then brightened. "Can I still wear the nose in the van?" Molly looked at him for a long moment. Eleven years of partnership sat between them like a sleeping dog. "In the van," she said, "you may wear the nose."
created_at
"2026-07-25T13:43:55.927Z"
7/25/2026, 1:43:55 PM
model
"anthropic/claude-opus-5-20260723"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
4379
moderation_latency
(null)
generation_time
23590
tokens_prompt
75
tokens_completion
824
native_tokens_prompt
93
native_tokens_completion
1174
native_tokens_completion_images
(null)
native_tokens_reasoning
68
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
0.029815
router
(null)
provider_responses
0
endpoint_id
"76cb4608-f48c-483d-8da8-9957fb44244e"
id
"msg_011CdNsi2bURCVb4fvqy7o5y"
is_byok
false
latency
1961
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-1784987035-cC1HsZKwjWlkXJtu1C7l"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784987035-tQz8rHr6cYKxZblCapq3"
upstream_id
"msg_011CdNsi2bURCVb4fvqy7o5y"
total_cost
0.029815
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
100.00%
Matches word count
n/a
neededClean
false
words
501
0.00%
Dialogue to Total Word Ratio
Ratio: 66.40%, Deviation: 16.40%
neededClean
false
wordsTotal
509
wordsDialogue
338
49.9996%