NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCQTD9400269HVP6NTWCCN9
Run Details
Model:
Claude Opus 5 (Reasoning, Low)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
11.2s
Cost:
$0.013165
Provider:
Amazon Bedrock
Evaluation summary
Score
Evaluator
Details Link
99.98%
Matches word count
(details)
51.74%
Dialogue to Total Word Ratio
(details)
75.8606%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 90% 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
"Right," said Molly. "Two o'clock, the bouncy castle goes up. That's our window." "Our window is a bouncy castle?" "The *distraction* is a bouncy castle, Bill. Keep up." "And the presents?" "Stacked on the trestle table by the gazebo. Forty-odd gift bags, minimum. Some of them have iPads in." "Seven-year-olds with iPads." "Don't get political, we're working." "I'm just saying, when I was seven I got a satsuma and a clip round the ear." "Which explains a great deal. Now. You'll be the entertainer." "The what?" "Mr. Bibbles. Balloon animals, magic tricks, the works. I've got the costume in the van." "Molly." "It's mostly polyester." "Molly, I can't do balloon animals." "You can do a sausage dog." "I can do a sausage." "That's a dog if you say it confidently." "And what are you doing while I'm humiliating myself in front of a horde of sugared-up infants?" "Loading the van. Twelve minutes, in and out, and we're eating steak by teatime." "What if a kid clocks me?" "They're seven. Deny everything." "That's your plan? Deny everything?" "It's held up in court twice, Bill." "Fine. But I'm not doing the voice." "You'll do the voice." "...I'll do the voice."
created_at
"2026-07-25T13:35:47.501Z"
7/25/2026, 1:35:47 PM
model
"anthropic/claude-opus-5-20260723"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
3982
moderation_latency
(null)
generation_time
11121
tokens_prompt
75
tokens_completion
329
native_tokens_prompt
93
native_tokens_completion
508
native_tokens_completion_images
(null)
native_tokens_reasoning
19
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
0.013165
router
(null)
provider_responses
0
endpoint_id
"76cb4608-f48c-483d-8da8-9957fb44244e"
id
"msg_011CdNs62k9MdoLBhUywBiJH"
is_byok
false
latency
2433
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-1784986547-xDyQcXhjxpLurskLxbiZ"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784986547-gtjAO7WFBhSFuvDmSGSE"
upstream_id
"msg_011CdNs62k9MdoLBhUywBiJH"
total_cost
0.013165
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
99.98%
Matches word count
n/a
neededClean
false
words
198
51.74%
Dialogue to Total Word Ratio
Ratio: 99.01%, Deviation: 9.01%
neededClean
false
wordsTotal
202
wordsDialogue
200
75.8606%