NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-10-09-01M4FXZN1SR6WTJRPPAZ8R4NYX
Run Details
Model:
Claude Haiku 5.5 (Reasoning, Low)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
5.7s
Cost:
$0.000369
Provider:
Claude Platform on AWS
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.07%
Dialogue to Total Word Ratio
(details)
0.0339%
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
Molly leaned over the sticky diner table and tapped the map with a french fry. "Okay, Bill. The Henderson birthday party starts at two. Eleven kids, one bouncy castle, and a mountain of presents in the living room." Bill squinted at the napkin sketch. "Why is the bouncy castle on the map? That's not a security feature." "It's a distraction. Nobody notices a burglar when there's a six-year-old flying off a trampoline." "Fair. What about the parents?" "Mr. Henderson is hiding in the garage with a sandwich and a podcast. Mrs. Henderson is in the kitchen, yelling about a cake that looks like a dinosaur. We're in and out before the candles get lit." Bill frowned. "I still think we should just buy presents and sell them back. Less paperwork." "Bill, we are not buying presents. We are stealing presents. That's the whole brand." "Our brand is 'mildly confused people in a van.'" "Our brand is professional." Molly pulled out two plastic masks shaped like clowns. "Wear these. Nobody suspects a clown." Bill held his up and stared into its painted grin. "Molly, I'm not wearing this. Kids are going to chase me." "Good. Then they'll stop looking for the robbers and start looking for the ice cream." "That's a terrible plan." "It's a great plan with a terrible name. Here's the route. We come through the side gate, grab the big boxes first, then the small shiny ones. Never take the wrapping paper. It's noisy." "What if a kid catches us?" Molly paused. "Then we give them a dollar and tell them the clown is a real magician." "We don't have a dollar." "Then we tell them the magician is very poor." Bill sighed and pushed his fries away. "Okay. But if I end up on a birthday cake, I'm blaming you." "Deal. Now eat your fries. We leave at one-fifty. Timing is everything." "How do you even know the party starts at two?" Molly smiled sweetly and slid a glittering invitation across the table. "Because I'm the one who was invited."
created_at
"2026-10-09T08:55:24.991Z"
10/9/2026, 8:55:24 AM
model
"anthropic/claude-haiku-5.5-20261007"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
733
moderation_latency
(null)
generation_time
5631
tokens_prompt
75
tokens_completion
499
native_tokens_prompt
95
native_tokens_completion
718
native_tokens_completion_images
(null)
native_tokens_reasoning
0
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
0.0003685
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1791536124-xjkGKKfwsE5FbGxZwdqb"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1791536124-8LK73IShbBnzo6rTd9m6"
upstream_id
"msg_011CfrPBYLYzne2y9NFW9AR4"
provider_responses
0
endpoint_id
"b6cdf493-3a61-441f-a134-e7944fddb980"
id
"msg_011CfrPBYLYzne2y9NFW9AR4"
is_byok
false
latency
733
model_permaslug
"anthropic/claude-haiku-5.5-20261007"
provider_name
"Claude Platform on AWS"
status
200
total_cost
0.0003685
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
343
0.07%
Dialogue to Total Word Ratio
Ratio: 82.08%, Deviation: 12.08%
neededClean
false
wordsTotal
346
wordsDialogue
284
0.0339%