NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-10-08-01M4DA1Z6WRCDA116VSSQ73N9R
Run Details
Model:
Claude Haiku 5.5 (Adaptive)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
10.3s
Cost:
$0.001019
Provider:
Claude Platform on AWS
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
99.65%
Dialogue to Total Word Ratio
(details)
49.8272%
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
The van smelled of cold fries and ambition. Molly tapped a blueprint spread across the dashboard while Bill squinted through the windshield at a bouncy castle. "Okay," Molly said. "Phase one. We walk in through the side gate." "The side gate has a sign that says Private Party, Adults Keep Out." "Which is why we're wearing these." She pulled two lanyards from her coat. Each card read CLOWN, ENTRY APPROVED. Bill took his. "Where'd you get clown passes?" "A man named Gary. Don't ask about Gary." "I'm asking about Gary." "Bill, focus. Phase two. The presents are stacked in the living room, behind the cake table." "How many presents?" "Forty-three, according to the mum's Facebook post." "Forty-three! What's in them?" "Mostly bikes, a drone, and a box Gary swears is a scooter." Bill frowned. "Molly, they're kids. Won't they notice their presents are gone?" "That's phase three, and it's the genius part. We leave each of them one present. Wrapped. Empty box. Little card that says, Sorry, Santa's lawyers have been notified." "That's not a real thing." "It is now. Nobody argues with a card that official." "What about the dog?" "What dog?" "The dog that's been following us since the car park." Molly turned slowly. A golden retriever wearing a party hat sat beside the bins, holding a squeaky duck in its mouth. "Oh, that's Biscuit," Bill said. "Birthday boy's dog. He's been helping himself to the sausage rolls." "Then he's an accomplice. Hand him a sausage roll, Bill." "Why me?" "Because you're good with animals." "I'm good with ferrets." "Ferrets are animals." Biscuit trotted over, dropped the duck at Bill's feet, and wagged. Bill sighed and produced a sausage roll from his coat pocket. "Where did you even get that?" "Don't ask about the sausage rolls either." Molly checked her watch. "Clown hour starts in four minutes. Remember, nobody touches the piñata." "Why is the piñata off limits?" "Because it's full of actual candy, and I have a moral code." Bill stared. "You have a moral code? Since when?" "Since the Hendersons' wedding." "Right. Let's not talk about the Hendersons." Biscuit barked once, sharp and approving. Molly grinned, grabbed a bag marked BALLOONS, and slid the door open. "Come on, partner. Let's go steal some birthdays."
created_at
"2026-10-08T08:28:40.549Z"
10/8/2026, 8:28:40 AM
model
"anthropic/claude-haiku-5.5-20261007"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
7821
moderation_latency
(null)
generation_time
10266
tokens_prompt
75
tokens_completion
593
native_tokens_prompt
95
native_tokens_completion
2019
native_tokens_completion_images
(null)
native_tokens_reasoning
1121
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
0.001019
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1791448120-3gz3lwHRn3gengrogOgr"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1791448120-BK5IMJXVwkk6e5uWyNoa"
upstream_id
"msg_011CfpTLMac3xGiVsFJu6Smo"
provider_responses
0
endpoint_id
"b6cdf493-3a61-441f-a134-e7944fddb980"
id
"msg_011CfpTLMac3xGiVsFJu6Smo"
is_byok
false
latency
802
model_permaslug
"anthropic/claude-haiku-5.5-20261007"
provider_name
"Claude Platform on AWS"
status
200
total_cost
0.001019
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
378
99.65%
Dialogue to Total Word Ratio
Ratio: 69.74%, Deviation: 0.26%
neededClean
false
wordsTotal
380
wordsDialogue
265
49.8272%