NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCQYWCRRQQJ56N47NNX4GH2
Run Details
Model:
Claude Opus 5 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
21.9s
Cost:
$0.030915
Provider:
Amazon Bedrock
Evaluation summary
Score
Evaluator
Details Link
93.65%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
46.8248%
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 of cold sausage rolls and ambition. Molly had the blueprints spread across the dashboard, though calling them blueprints was generous; it was a leaflet for Ferndale Community Hall with arrows drawn on it in purple crayon. "Right," she said. "Fire exit here. Kitchen hatch here. Presents will be stacked on the trestle table by the window, because that's where mums always stack them." Bill squinted at the leaflet. "Why is the crayon purple?" "Because that's the only one Sasha wasn't using." "You borrowed crayons from your niece to plan a robbery." "I borrowed crayons from my niece to plan a *liberation*. There's a difference and it's mostly legal." "There is no version of this that's legal, Moll. It's a seventh birthday party." "It's forty-two gifts, unopened, still in wrapping, no serial numbers, no receipts, no insurance claims, and no witnesses over four feet tall." She tapped the leaflet. "It's the softest target in the county." Bill rubbed his face with both hands, slowly, like a man trying to erase himself. "Children have very good memories." "Children have *terrible* memories. Ask a child what happened yesterday. They'll tell you about a dog that could talk." "Ask a child what the bad man looked like and they'll say 'big sad bald one with the eyebrows,' and then they'll draw you, Bill, in crayon, and the police will laminate it." "That's why you're wearing the costume." There was a pause of considerable weight. "What costume." Molly reached into the back and produced, with the reverence of a knight presenting a sword, a bin bag containing something enormous, yellow, and matted. "Absolutely not." "It's a bee." "It's a *diseased* bee." "It's an authentic children's entertainer bee, and it cost me eleven pounds, and it means nobody looks at your face, and it means when you walk out of that hall with a sack over your shoulder, forty-two children applaud you." Molly spread her hands. "It's not a getaway. It's an *encore*." Bill stared into the bin bag for a long moment. "What are you going as?" "Aunt Deborah." "Whose Aunt Deborah?" "Everyone's. Every party has an Aunt Deborah and nobody knows whose she is. She stands near the sandwiches, she says 'they grow up so fast,' and she is invisible to man and God." Molly pulled a floral cardigan from under her seat and shrugged it on like body armour. "I've been Aunt Deborah at nineteen christenings." "And the sack?" "Bin bags. Nobody questions an adult tidying at a party. You could carry a body out of a soft play centre in a bin bag if you looked tired enough." Bill exhaled through his nose. "And the plan is what, precisely? Timings." "Two-fifteen, cake. Everyone faces the cake. Two-sixteen, lights off for candles. Two-sixteen and thirty seconds, you're at the trestle table. Two-seventeen, they're singing, which covers the wrapping paper noise. Two-eighteen, fire exit, van, gone." "And if something goes wrong?" Molly considered this, then handed him the bee head. "Then you improvise a dance, Bill. Bees are very forgiving."
created_at
"2026-07-25T13:38:14.048Z"
7/25/2026, 1:38:14 PM
model
"anthropic/claude-opus-5-20260723"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
3190
moderation_latency
(null)
generation_time
21829
tokens_prompt
75
tokens_completion
850
native_tokens_prompt
93
native_tokens_completion
1218
native_tokens_completion_images
(null)
native_tokens_reasoning
69
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
0.030915
router
(null)
provider_responses
0
endpoint_id
"76cb4608-f48c-483d-8da8-9957fb44244e"
id
"msg_011CdNsGpCPuLojHmZWwSHba"
is_byok
false
latency
1515
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-1784986694-Da6tvh5APllup4Nfeqj2"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784986694-V6Cksli2z6g3srwTbZby"
upstream_id
"msg_011CdNsGpCPuLojHmZWwSHba"
total_cost
0.030915
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
93.65%
Matches word count
n/a
neededClean
false
words
509
0.00%
Dialogue to Total Word Ratio
Ratio: 72.67%, Deviation: 22.67%
neededClean
false
wordsTotal
516
wordsDialogue
375
46.8248%