NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-10-08-01M4DANZZTA5CFFBRRNNPDACFD
Run Details
Model:
Claude Haiku 5.5 (Adaptive)
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
9.3s
Cost:
$0.000996
Provider:
Claude Platform on AWS
Evaluation summary
Score
Evaluator
Details Link
99.74%
Matches word count
(details)
99.83%
Dialogue to Total Word Ratio
(details)
99.7855%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 50% 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 three houses down from the birthday party, its headlights off. Through the windshield, the backyard glowed with fairy lights. Somewhere inside, a hundred pounds of sugar was being shoved into toddlers, and a dozen wrapped presents sat in a glittering pile by the gate. Molly tapped the blueprint spread across the dashboard. "Right, Bill. Entry through the bouncy castle. Nobody checks a bouncy castle." "Got it. And the guard?" "Guard's a clown named Doodles. He's been juggling since noon. We wait until he hits the cake table." Bill squinted at the sketch, then at the small X drawn beside the piƱata. "What's this?" "Security camera. Blind spot, supposedly." "Supposedly?" "The kid sitting under it was in the magic show. His eyes were on the rabbit, not the camera." Bill adjusted his fake beard. It itched terribly, and the glue smelled like burnt toffee. He nodded slowly. "And the getaway?" "Ice cream truck. Playing that song, so nobody looks twice." "Molly, I'm nervous. There are forty children in there." Molly grinned, sharp as a paper cut. "Forty witnesses who can't pick a suspect out of a lineup. Trust me. Now put on the mustache."
created_at
"2026-10-08T08:39:36.704Z"
10/8/2026, 8:39:36 AM
model
"anthropic/claude-haiku-5.5-20261007"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
7911
moderation_latency
(null)
generation_time
9275
tokens_prompt
75
tokens_completion
300
native_tokens_prompt
95
native_tokens_completion
1972
native_tokens_completion_images
(null)
native_tokens_reasoning
1520
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
0.0009955
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1791448776-ljN0M62D0K8rpSLlSN5m"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1791448776-aA8Ruo85GgIPiJ4VFT7K"
upstream_id
"msg_011CfpUAintuv5sJ7JjgXBRp"
provider_responses
0
endpoint_id
"b6cdf493-3a61-441f-a134-e7944fddb980"
id
"msg_011CfpUAintuv5sJ7JjgXBRp"
is_byok
false
latency
701
model_permaslug
"anthropic/claude-haiku-5.5-20261007"
provider_name
"Claude Platform on AWS"
status
200
total_cost
0.0009955
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
99.74%
Matches word count
n/a
neededClean
false
words
196
99.83%
Dialogue to Total Word Ratio
Ratio: 52.04%, Deviation: 2.04%
neededClean
false
wordsTotal
196
wordsDialogue
102
99.7855%