NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCQRABPAZRYEBWBD9GK52QR
Run Details
Model:
Claude Opus 5 (Reasoning, Low)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
21.6s
Cost:
$0.028490
Provider:
Amazon Bedrock
Evaluation summary
Score
Evaluator
Details Link
98.71%
Matches word count
(details)
0.01%
Dialogue to Total Word Ratio
(details)
49.3593%
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 professional enough, given the target. "Right," she said, tapping a greasy corner with a pen. "Bouncy castle here. Trestle table with the sandwiches here. Presents stacked in the conservatory, unguarded except for a nan." Bill leaned over the box, chewing his thumbnail. "How big's the nan?" "Does it matter?" "Course it matters. Nans come in sizes. There's the little wispy ones that fall asleep in the armchair, and then there's the other kind. The kind with a handbag full of pound coins and no fear of death." "She's about five foot two." "That's the dangerous height," said Bill grimly. "That's the height they're all made at." Molly sat back and folded her arms. In eleven years of crime, she had robbed a bookmaker's, three off-licences, and a man dressed as a wizard, and she had never once had to discuss a nan's dimensions in the planning stage. "The party starts at two," she said. "Fifteen kids, one hired clown, one bouncy castle. Parents will be in the kitchen drinking wine and pretending they don't hate each other. Nobody's watching the loot." "And we just walk in?" "We're invited." Bill's face did something complicated. "We're what?" "I told the mother we were the entertainment. Balloon animals." "Molly." "What?" "I can't do balloon animals." "Nobody can do balloon animals. You twist it twice and call it a dog. Children have no standards, that's the beauty of them. A sausage with an ear is a triumph." "And what are you doing while I'm ruining balloons?" "Loading the presents into the van through the conservatory door. Should take four minutes." Molly slid a paper bag across the table. "Wear this." Bill opened it and pulled out a rainbow wig and a red foam nose. He held them at arm's length like a dead pigeon. "I want it on record," he said, "that this is beneath us." "You stole a Wendy house last April." "That was an opportunity. There was a Wendy house and there was nobody standing near it. This is a plan. Plans have dignity requirements." "Put the nose on, Bill." He put the nose on. He looked, if anything, less trustworthy than before, which Molly hadn't thought was geometrically possible. "One rule," she said. "No sweets." "I wasn't going to take sweets." "You took sweets at the christening job." "I took *one* bag of Haribo and I've apologised to you at length—" "You sat down in the car and ate them and then you couldn't drive because you said your hands felt fizzy." "That was a medical event," said Bill, with what remained of his dignity. Molly stood, folded the pizza box, and shoved it in the bin. Outside, the van was idling badly, as it had for six years. "Two o'clock," she said. "In and out. Fifteen presents, no casualties, no sweets." "And the nan?" "If the nan moves," Molly said, opening the door, "you make her a balloon dog and you run."
created_at
"2026-07-25T13:34:38.974Z"
7/25/2026, 1:34:38 PM
model
"anthropic/claude-opus-5-20260723"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
3713
moderation_latency
(null)
generation_time
21463
tokens_prompt
75
tokens_completion
794
native_tokens_prompt
93
native_tokens_completion
1121
native_tokens_completion_images
(null)
native_tokens_reasoning
48
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
0.02849
router
(null)
provider_responses
0
endpoint_id
"76cb4608-f48c-483d-8da8-9957fb44244e"
id
"msg_011CdNs11pwS616cYkwHVKhR"
is_byok
false
latency
2177
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-1784986478-F2OKZIuNP84XRo5NDXOK"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784986478-NKGIeRIH4AQZZOoA5M34"
upstream_id
"msg_011CdNs11pwS616cYkwHVKhR"
total_cost
0.02849
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
98.71%
Matches word count
n/a
neededClean
false
words
506
0.01%
Dialogue to Total Word Ratio
Ratio: 63.91%, Deviation: 13.91%
neededClean
false
wordsTotal
507
wordsDialogue
324
49.3593%