NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-10-08-01M4DACDZKVQGSYXTPQ9HC8K9T
Run Details
Model:
Claude Haiku 5.5 (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
10.5s
Cost:
$0.000978
Provider:
Claude Platform on AWS
Evaluation summary
Score
Evaluator
Details Link
6.09%
Matches word count
(details)
99.34%
Dialogue to Total Word Ratio
(details)
52.7133%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 10% 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 getaway van idled behind a bouncy castle warehouse, headlights off, engine humming like a nervous bee. Molly sat in the passenger seat, tapping a clipboard against her knee, while Bill squinted through binoculars at the party across the street. Pastel balloons bobbed above the garden fence. Somewhere inside, a clown was murdering a ukulele version of a birthday song. "Forty-two presents," Bill whispered. "Zero witnesses." Molly flipped a page on her clipboard, its corners already stained with frosting. Her plan involved a fake mustache, a stolen tray of cupcakes, and a magician who had no idea he was about to be distracted by her. "And no crumbs," she said. Bill lowered the binoculars and sighed. On their last heist, they had left a trail of sprinkles across three counties, and the police had followed it straight to their front door. Molly pulled on the mustache, which immediately slid sideways like a tired caterpillar. She pressed it harder and tried again, wincing as it stuck to her upper lip with a sound like tape ripping off a box. Bill reached into his coat, unwrapped a lollipop, and popped it into his mouth. "Relax," he said around the candy. "Nobody suspects two uncles with sticky hands." Molly adjusted her wig, checked the clock on the dashboard, and grinned. Operation Gift Grab was officially live.
created_at
"2026-10-08T08:34:23.356Z"
10/8/2026, 8:34:23 AM
model
"anthropic/claude-haiku-5.5-20261007"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2393
moderation_latency
(null)
generation_time
10464
tokens_prompt
75
tokens_completion
510
native_tokens_prompt
95
native_tokens_completion
1936
native_tokens_completion_images
(null)
native_tokens_reasoning
1463
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
0.0009775
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1791448463-g6NYnxiBdcurUx0lq3TB"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1791448463-GLjTL5W09Rzh8YCuSI7L"
upstream_id
"msg_011CfpTmdD9v8PCsQpu1A6CE"
provider_responses
0
endpoint_id
"b6cdf493-3a61-441f-a134-e7944fddb980"
id
"msg_011CfpTmdD9v8PCsQpu1A6CE"
is_byok
false
latency
783
model_permaslug
"anthropic/claude-haiku-5.5-20261007"
provider_name
"Claude Platform on AWS"
status
200
total_cost
0.0009775
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
6.09%
Matches word count
n/a
neededClean
false
words
223
99.34%
Dialogue to Total Word Ratio
Ratio: 7.14%, Deviation: 2.86%
neededClean
false
wordsTotal
224
wordsDialogue
16
52.7133%