NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-10-08-01M4E1G2Y010X43TP3Z26KRDPD
Run Details
Model:
Claude Haiku 5.5 (Reasoning, Low)
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
13.3s
Cost:
$0.001317
Provider:
Claude Platform on AWS
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
100.00%
Dialogue to Total Word Ratio
(details)
49.9994%
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 van idled behind the bouncy castle warehouse, headlights off, engine sputtering like a nervous hamster. Molly spread the blueprints across the dashboard and traced the route with a gloved finger, her ski mask itching in the summer heat. Beside her, Bill hunched over a pair of plastic binoculars he had bought at the party store, convinced they made him look professional. Molly leaned in and whispered the details. "The party's in the east room. Eleven kids, one clown, and a unicorn piƱata." She added that the gift table sat beside the cake, so they needed to grab every box with a bow and be gone before anyone noticed. Bill squinted at the plans, then at his reflection in the rearview mirror. He had cut the eye holes in his mask at the wrong height, so he could see only the floor and the bottom of his nose. He adjusted it twice, which made things worse. "Why do they need a unicorn?" he asked. Molly sighed and tapped her grappling hook, which was really a kitchen whisk duct-taped to a garden hose. Over the hedge, a child shrieked with joy, and both thieves froze, clutching their pillowcases for loot. "Nobody gets hurt," Molly reminded him, sliding her glittery squirt gun into her belt. "We're professionals." Bill nodded solemnly, climbed out of the van, bumped his head on the mailbox, and stepped directly into the flowerbed.
created_at
"2026-10-08T15:18:20.363Z"
10/8/2026, 3:18:20 PM
model
"anthropic/claude-haiku-5.5-20261007"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2647
moderation_latency
(null)
generation_time
13270
tokens_prompt
75
tokens_completion
646
native_tokens_prompt
95
native_tokens_completion
2615
native_tokens_completion_images
(null)
native_tokens_reasoning
2138
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
0.001317
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1791472700-FTaMOxhafovEoptsBpbj"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1791472700-g3bKZZeyCyndgwGfp3FJ"
upstream_id
"msg_011CfpzaFAQAWQgPC1pg35Wq"
provider_responses
0
endpoint_id
"b6cdf493-3a61-441f-a134-e7944fddb980"
id
"msg_011CfpzaFAQAWQgPC1pg35Wq"
is_byok
false
latency
817
model_permaslug
"anthropic/claude-haiku-5.5-20261007"
provider_name
"Claude Platform on AWS"
status
200
total_cost
0.001317
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
235
100.00%
Dialogue to Total Word Ratio
Ratio: 10.59%, Deviation: 0.59%
neededClean
false
wordsTotal
236
wordsDialogue
25
49.9994%