NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-10-08-01M4DA8E1ME1NSW0W0K5TC80N6
Run Details
Model:
Claude Haiku 5.5 (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
8.4s
Cost:
$0.000713
Provider:
Claude Platform on AWS
Evaluation summary
Score
Evaluator
Details Link
14.30%
Matches word count
(details)
26.69%
Dialogue to Total Word Ratio
(details)
20.4944%
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 Hendersons' garage, headlights off, its engine coughing like a grandfather at a funeral. Molly spread a blueprint across the dashboard while Bill chewed a sandwich he had clearly swiped from the party's catering table. "Presents are in the living room," Molly whispered. "Eight, maybe nine." Bill swallowed. "What about the bouncy castle?" Outside, a balloon drifted past the window, bobbing against the streetlight like a tiny, cheerful ghost. Somewhere inside the house, a child shrieked with joy, and both thieves flinched in unison. Molly pulled a ski mask over her curls and checked the tool bag. Glitter spilled onto the seat. Crowbar, zip ties, and a rubber duck she didn't remember packing. "We're stealing gifts, not souls," she said. "Keep up." Bill brushed crumbs from his coat. "Those boxes have puppies printed on them, Molly." She ignored him and eased open the door. Cold night air rushed in, carrying the smell of birthday cake and the distant bleat of a kazoo. A sprinkler ticked across the lawn, and a half-inflated unicorn sagged near the fence like it had given up on life. They crept toward the back gate, stepping carefully past a wobbling trampoline. Bill stopped to pocket a cupcake off a picnic table. "Mission first," Molly hissed. "Morale first," Bill replied, and bit into it.
created_at
"2026-10-08T08:32:12.348Z"
10/8/2026, 8:32:12 AM
model
"anthropic/claude-haiku-5.5-20261007"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
3359
moderation_latency
(null)
generation_time
8349
tokens_prompt
75
tokens_completion
427
native_tokens_prompt
95
native_tokens_completion
1407
native_tokens_completion_images
(null)
native_tokens_reasoning
903
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
0.000713
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1791448332-bEel2SX5c54cNKbufOWV"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1791448332-7nUToL4dBVcEoaYZKaCK"
upstream_id
"msg_011CfpTbyCAHiVrSGfzcK6K8"
provider_responses
0
endpoint_id
"b6cdf493-3a61-441f-a134-e7944fddb980"
id
"msg_011CfpTbyCAHiVrSGfzcK6K8"
is_byok
false
latency
835
model_permaslug
"anthropic/claude-haiku-5.5-20261007"
provider_name
"Claude Platform on AWS"
status
200
total_cost
0.000713
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
14.30%
Matches word count
n/a
neededClean
false
words
221
26.69%
Dialogue to Total Word Ratio
Ratio: 20.72%, Deviation: 10.72%
neededClean
false
wordsTotal
222
wordsDialogue
46
20.4944%