NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-10-09-01M4FZAAZ8J1FN8GV2WNEFR8V9
Run Details
Model:
Claude Haiku 5.5 (Reasoning, Low)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
3.3s
Cost:
$0.000215
Provider:
Claude Platform on AWS
Evaluation summary
Score
Evaluator
Details Link
14.30%
Matches word count
(details)
99.44%
Dialogue to Total Word Ratio
(details)
56.8707%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 90% 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
"Okay, Bill, go over it one more time," Molly whispered, tapping the blueprint of the Pembroke house. "Who's guarding the presents?" "Nobody. Just a clown named Sparkles," Bill said. "He's got a bow tie and a fake nose." "A clown? Bill, clowns are the most dangerous people alive. Their smiles are painted on." "He's seven feet tall and wears no shoes, Molly." "That's a warning sign!" "Fine. We'll go in through the chimney," Bill said. "Santa's suit is in the van. Put it on." "I'm not wearing a red suit with a fake beard, Bill. I look ridiculous." "You'll blend in. Nobody expects a tiny elf thief." "Elf? You said Santa!" "Santa's busy. We're the elves now." "And what's the plan once we're inside?" "We grab the presents, load them into the sack, and vanish before the cake comes out." "What if a kid wakes up?" "Then we tell them the tooth fairy sent us, and they'll hand over their own birthday loot." "Bill, that's the worst plan I've ever heard." "Thanks. I've been working on it all week."
created_at
"2026-10-09T09:18:43.694Z"
10/9/2026, 9:18:43 AM
model
"anthropic/claude-haiku-5.5-20261007"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
3291
moderation_latency
(null)
generation_time
3291
tokens_prompt
75
tokens_completion
262
native_tokens_prompt
95
native_tokens_completion
411
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
0.000215
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1791537523-OhiAjd3YRk5HdAUllXjy"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1791537523-0CAsxT6GH1Fb3BkflrJm"
upstream_id
"msg_011CfrQxeNZhjUtnBaurYNhW"
provider_responses
0
endpoint_id
"b6cdf493-3a61-441f-a134-e7944fddb980"
id
"msg_011CfrQxeNZhjUtnBaurYNhW"
is_byok
false
latency
772
model_permaslug
"anthropic/claude-haiku-5.5-20261007"
provider_name
"Claude Platform on AWS"
status
200
total_cost
0.000215
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
179
99.44%
Dialogue to Total Word Ratio
Ratio: 92.74%, Deviation: 2.74%
neededClean
false
wordsTotal
179
wordsDialogue
166
56.8707%