NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-10-08-01M4DAD5ZHYKHNSEBTN519XCAW
Run Details
Model:
Claude Haiku 5.5 (Adaptive)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
12.2s
Cost:
$0.001234
Provider:
Claude Platform on AWS
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.02%
Dialogue to Total Word Ratio
(details)
0.0090%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 30% 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 community hall smelled of balloons, sticky frosting, and the desperation of a clown who had been paid far less than he deserved. Molly crouched behind a hedge across the street, binoculars pressed to her eyes. Bill hunched beside her in a stolen Santa coat two sizes too small, already missing a button somewhere near his stomach. "Tell me again why we're robbing a birthday party," Bill whispered. "Because it's a birthday party, Bill. Forty-two kids, mountains of gifts, and not one adult paying attention because they're all busy arguing about whose kid bit whose kid." "Sounds ethical." "It's the most ethical heist in town. Nobody's going to miss a Lego set they never opened." Bill squinted at the hall. Through the window, a man in a rainbow wig was juggling bowling pins while a row of toddlers shrieked with delight. One of them was licking the cake off the ceiling fan. "What about the clown?" "Gerald? He's been juggling the same three pins since the nineties. He won't notice a thing." "What about the bouncy castle? That thing would fetch a fair price." "Bill, we are not stealing the castle. We're stealing presents. Small, wrapped, shiny presents. Focus." Bill adjusted his fake beard, which slid sideways and dangled over one ear. He pushed it back into place with the dignity of a man who had never once been embarrassed by facial hair, and failed. Molly sighed and checked her watch. "Phase one, you go in the side door pretending to be the pizza delivery guy. Phase two, I follow as the aunt who forgot the candles." "And if someone asks questions?" "We cry. Nobody questions a crying aunt." Bill grinned. "You're a genius." "I'm a professional." She pulled a plastic bag from her coat and shook it open with a crisp snap. "Now, the presents with the glitter bows go first. Those are the expensive ones." "How do you know?" "Glitter costs money. Rich parents always overcompensate." Across the street, the side door creaked open. A teenager in a grease-stained apron stepped out, lit a cigarette, and promptly dropped a pizza box into the gutter. Bill straightened his coat, took a deep breath, and nodded solemnly. "Right. Let's go be terrible people." Molly stood, brushed off her knees, and started across the road. Behind them, Gerald the clown finally dropped all three pins at once, and somewhere a child began to cry, which, Molly thought happily, would make her job even easier.
created_at
"2026-10-08T08:34:47.93Z"
10/8/2026, 8:34:47 AM
model
"anthropic/claude-haiku-5.5-20261007"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
12214
moderation_latency
(null)
generation_time
12214
tokens_prompt
75
tokens_completion
618
native_tokens_prompt
95
native_tokens_completion
2450
native_tokens_completion_images
(null)
native_tokens_reasoning
1588
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
0.0012345
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1791448487-TmaVAXUMzZ6ozWfcnWAZ"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1791448487-6kKc1L826xraifKSks1A"
upstream_id
"msg_011CfpToS8TXtpwcZqkjvUTz"
provider_responses
0
endpoint_id
"b6cdf493-3a61-441f-a134-e7944fddb980"
id
"msg_011CfpToS8TXtpwcZqkjvUTz"
is_byok
false
latency
728
model_permaslug
"anthropic/claude-haiku-5.5-20261007"
provider_name
"Claude Platform on AWS"
status
200
total_cost
0.0012345
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Claude Platform on AWS"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
413
0.02%
Dialogue to Total Word Ratio
Ratio: 43.13%, Deviation: 13.13%
neededClean
false
wordsTotal
415
wordsDialogue
179
0.0090%