NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCRFRX5FRK7RV6GP8S00XVQ
Run Details
Model:
Claude Opus 5 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
10.6s
Cost:
$0.013765
Provider:
Amazon Bedrock
Evaluation summary
Score
Evaluator
Details Link
68.10%
Matches word count
(details)
50.64%
Dialogue to Total Word Ratio
(details)
59.3690%
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, so the target is a seven-year-old," Molly said. "Say it louder, why don't you." "It's a bouncy castle, Bill, not Fort Knox." "That bouncy castle has a *perimeter*." "It has a deflated corner and a sign that says NO SHOES." "I'm just saying, we go in through the garden gate, we get the presents, we're out before the cake." "Before the cake?" "Before the cake." "Bill, we're not leaving before the cake." "That's not—Molly, that's not how a heist works." "It's a *victoria sponge*. Denise made it. Denise has won things." "You know Denise?" "Denise is my cousin." "Your cousin is hosting the party we are robbing?" "Which is how I know there's a Nintendo in the pile. Wrapped in penguins." "So you're on the guest list." "I'm on the guest list, I brought a gift bag, and I'm doing the pass-the-parcel music." "Then what do you need me for?" "You're the muscle." "Against who? Toddlers?" "Against Gary." "Who's Gary?" "Gary's the dad. Gary does CrossFit and he *watches* the present table." "So distract Gary." "That's your job. Ask him about his deadlift." "And then?" "And then he'll talk for forty minutes and I'll be in the car with a Nintendo and two slices of sponge." "...Three slices." "Don't be greedy, Bill."
created_at
"2026-07-25T13:47:27.532Z"
7/25/2026, 1:47:27 PM
model
"anthropic/claude-opus-5-20260723"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2472
moderation_latency
(null)
generation_time
10477
tokens_prompt
75
tokens_completion
349
native_tokens_prompt
93
native_tokens_completion
532
native_tokens_completion_images
(null)
native_tokens_reasoning
21
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
0.013765
router
(null)
provider_responses
0
endpoint_id
"76cb4608-f48c-483d-8da8-9957fb44244e"
id
"msg_011CdNsycvsWa8CHGBXpRjFh"
is_byok
false
latency
1402
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-1784987247-l0KrL8IcWDfP8HxPr9ow"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784987247-6hEKvSnkaXzVHczYNNtn"
upstream_id
"msg_011CdNsycvsWa8CHGBXpRjFh"
total_cost
0.013765
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
68.10%
Matches word count
n/a
neededClean
false
words
214
50.64%
Dialogue to Total Word Ratio
Ratio: 99.08%, Deviation: 9.08%
neededClean
false
wordsTotal
218
wordsDialogue
216
59.3690%