NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCQPRDS9ZP8ADWJM8QC830E
Run Details
Model:
Claude Opus 5 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
24.2s
Cost:
$0.030940
Provider:
Amazon Bedrock
Evaluation summary
Score
Evaluator
Details Link
95.99%
Matches word count
(details)
3.21%
Dialogue to Total Word Ratio
(details)
49.5966%
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, smoothing it with her palm. "That's a napkin," Bill said. "It's a schematic." "It's a napkin from Gino's, Molly. There's a marinara stain where the back door should be." "That's the bouncy castle. Now focus." She tapped the napkin. "Fourteen presents. Maybe sixteen if the Delgado twins overachieve, which they will, because Marla Delgado has something to prove ever since the christening incident." "And the target?" "Sophie Pemberton. Turning seven. Party starts at two, cake at three, presents opened at four." "So we go in at three-fifty." "We go in at three-fifteen, during cake, when the parents are filming and the kids are sugar-blind." Molly folded the napkin and tucked it into her jacket. "You'll be the entertainment." Bill stopped chewing his gum. "Define entertainment." "Clown." "No." "Bill." "I said no. We agreed after Reno. No costumes with pockets I can't reach." "You'll have a whole trunk. Clowns get trunks. Nobody questions a clown with a trunk, it's practically expected." She opened the van's back doors, revealing a rack of polyester horrors. "I got you the yellow one. It's slimming." "It has a ruffle." "All the good ones have a ruffle." Bill held up the costume against himself, grimly. "And what are you doing while I'm honking a bicycle horn for a room full of feral seven-year-olds?" "Aunt Deborah." "Whose aunt?" "Nobody's. That's the beauty of Aunt Deborah. Every family has one and nobody admits they can't place her. I'll bring a casserole and complain about my hip until they beg me to sit somewhere out of the way." She mimed a limp, wincing theatrically. "Preferably near the gift table." "You've thought about this a disturbing amount." "I've been in the game nineteen years, Bill. I've never once been caught by a woman named Deborah and neither has anybody else." She pulled a canvas tote from the van. "Presents go in here, tote goes in your trunk, trunk goes out the side gate at three-forty." "What about the dog?" Molly paused. "What dog?" "The Pembertons have a dog. Big one. Brown." "How do you know that?" "Because I did surveillance, Molly, unlike some people who did their surveillance on a napkin." Bill dug in his pocket and produced a photograph. "That's Bartholomew. Ninety pounds. Steals sandwiches." "Then he's a colleague." She took the photo, studying it. "Fine. Bring hot dogs." "How many?" "How many do you think?" "I don't know how many hot dogs a ninety-pound dog needs, I'm a thief, not a—" "Bring the whole pack." Bill exhaled and pulled the clown wig over his head, where it sat like a startled poodle. "And if a kid asks me to make a balloon animal?" "Make a balloon animal." "I can't make a balloon animal." "Then make a balloon snake and tell them it's a very sad giraffe." "That's cruel." "That's improvisation." Molly slammed the doors and checked her watch. "It's one-fifty. Get in the van, Bartholomew's counting on us." "Bartholomew's the dog." "I know what I said."
created_at
"2026-07-25T13:33:47.983Z"
7/25/2026, 1:33:47 PM
model
"anthropic/claude-opus-5-20260723"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
3903
moderation_latency
(null)
generation_time
23849
tokens_prompt
75
tokens_completion
830
native_tokens_prompt
93
native_tokens_completion
1219
native_tokens_completion_images
(null)
native_tokens_reasoning
50
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
0.03094
router
(null)
provider_responses
0
endpoint_id
"76cb4608-f48c-483d-8da8-9957fb44244e"
id
"msg_011CdNrwDvbpD7vVnWVfEGfS"
is_byok
false
latency
2210
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-1784986428-lT4LZPCWoZ3uxr2aSsF2"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784986428-taduwEaM17ZkDNt8uPVN"
upstream_id
"msg_011CdNrwDvbpD7vVnWVfEGfS"
total_cost
0.03094
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
3.21%
Dialogue to Total Word Ratio
Ratio: 78.29%, Deviation: 8.29%
neededClean
false
wordsTotal
516
wordsDialogue
404
49.5966%