NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-05-27-01KSMF0H0BFDZJRXXC77JW449G
Run Details
Model:
Cydonia 24B V4.1
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
14.3s
Cost:
$0.000370
Provider:
Parasail
Evaluation summary
Score
Evaluator
Details Link
75.16%
Matches word count
(details)
9.71%
Dialogue to Total Word Ratio
(details)
42.4331%
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 flattened herself against the side of the bouncy castle, her dark hair bobbing against the plastic wall as she tried to peek inside. "Bill, I've gotCookie Monster on visual. He's got at least three presents already stuffed in that giant backpack of his." Bill adjusted his Groucho Marx glasses as he scammed the perimeter, holding a brightly colored party invitation in one hand. "At my age, I never thought I'd be breaking into a five-year-old's birthday party. The platforms on our getaway scooters must be slowing my reflexes." "It's an economic downturn, Bill. Children's birthday parties nowadays have better security than the Federal Reserve." Molly wiped sweat from her brow under the bright afternoon sun. "This clown is clearly an off-duty SWAT team member. Watch how he squeezes that balloon animal. Professional-grade vascular development." Bill nudged her forward. "The target is on the other side of the cake smashing ceremony. We go in during the chaos, extract the presents, and disappear into the neighborhood. Simple." "Simple is what the people who get caught always say." Molly rolled her eyes. "There's a security upgrade since we scouted. They've installed children's Disney themed juice boxes in those clear containers. Probably connected to some early warning system." "Are you saying juice boxes are going to be the downfall of our heist?" "I'm saying Minnie Mouse calling the police because her magic elixir has been disturbed is a legitimate operational threat." Molly reached into her fanny pack labeled "Craft Supplies" and pulled out a small, electronic device. "I've bypassed the parental control on the iPad running the music. When the kids start singing ' Cake By the Ocean,' that's our signal." "But that's not how the song goes." "It is today, Bill." Molly pocketed the device. "Now, wipe that suspicious smile off your face. Everyone knows middle-aged male thieves can't maintain a convincing 'I belong here' expression." Bill forced a grin. "Wouldn't want to stand out in a crowd of sandcastle enthusiasts and face painters." "Exactly. And whatever you do, don't try to blend in by eating those little cheese nuggets. Previous operations show that you can never eat just one of those things." "Copy that. Operation Cheese Nugget Avoidance: activated." Bill took a step forward, nearly tripping over a small plastic machine labeled " Instant Mini Donuts." "These kids have everything now. Mini donut machines on demand. Back in my day, we were lucky if we got a piñata filled with disappointment." "They've probably improved piñatas too. Now they come equipped with GPS tracking and facial recognition software," Molly deadpanned. "Current operational status?" Bill asked as music began to play in the distance. "Status of excessive parental units. Current count: seven. jumping castle attendance: high quality. Passive children's toy deployment: moderate. Congratulations, Bill. We're about to infiltrate a celebration of economic opportunity designed specifically to bankrupt the parents." Molly took a deep breath. "On your mark." "Mark." "Go team." "Go team."
created_at
"2026-05-27T10:17:17.078Z"
5/27/2026, 10:17:17 AM
model
"thedrummer/cydonia-24b-v4.1"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
220
moderation_latency
(null)
generation_time
14146
tokens_prompt
77
tokens_completion
808
native_tokens_prompt
243
native_tokens_completion
667
native_tokens_completion_images
(null)
native_tokens_reasoning
0
native_tokens_cached
240
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
"stop"
service_tier
(null)
usage
0.0003704
router
(null)
provider_responses
0
endpoint_id
"4d46db86-63fe-47eb-85f1-6a7567af190c"
id
"chatcmpl-45ebcc66f50841bc8dc5a18556c0826a"
is_byok
false
latency
220
model_permaslug
"thedrummer/cydonia-24b-v4.1"
provider_name
"Parasail"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.15.0; linux; x64))"
http_referer
(null)
request_id
"req-1779877037-VhWJ4GhXSALYbrcJvaOj"
session_id
(null)
api_type
"completions"
id
"gen-1779877037-XJbKygWu1cT2bcHfP1Wf"
upstream_id
"chatcmpl-45ebcc66f50841bc8dc5a18556c0826a"
total_cost
0.0003704
cache_discount
0.000036
upstream_inference_cost
0
provider_name
"Parasail"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
75.16%
Matches word count
n/a
neededClean
false
words
487
9.71%
Dialogue to Total Word Ratio
Ratio: 76.83%, Deviation: 6.83%
neededClean
false
wordsTotal
492
wordsDialogue
378
42.4331%