NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-17-01KHNF0M7P65SEP325TTCBHD4B
Run Details
Model:
Arcee AI: Trinity Mini
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
7.3s
Cost:
$0.000241
Provider:
Clarifai
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.25%
Dialogue to Total Word Ratio
(details)
0.1225%
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 leaned against the wall, arms crossed, eyes gleaming with mischief. "Bill, my dear, we're not just robbing a house. We're hitting the jackpot. The ultimate score. The children's party." Bill adjusted his glasses, looking skeptical. "Molly, we've discussed this. Stealing presents from a birthday bash? It's... logistically challenging. And ethically questionable, even for us." "Logistically challenging?" Molly scoffed. "Pfft. It's a walk in the park. Think about it. Parents are distracted. Kids are... well, kids. Mostly occupied with cake and screaming. The presents are just sitting there, under the tree, waiting to be... liberated." "Waiting to be stolen," Bill corrected, a frown forming. "And how exactly do you propose we get past the parents? Or the clown? Or the... the bouncy castle guard?" Molly waved a dismissive hand. "The parents? They're harmless. Mostly just trying to look like they're having fun. We blend in. We act like we belong. Maybe we pose as... I don't know... the 'gift delivery guys' the parents hired? Everyone loves a delivery guy, right?" Bill rubbed his temples. "Molly, the parents didn't hire anyone. And the clown... he's got a whistle. And a red nose. And he's probably armed with confetti cannons." "Confetti cannons are just... loud noise makers," Molly argued. "We'll be long gone before he even gets the first one out. And the bouncy castle guard? He's probably just a dad in a jumpsuit. We can... negotiate. Offer him a share. He looks like he could use the cash." Bill sighed, leaning against the wall opposite her. "Negotiate? With a bouncy castle guard? Molly, this is madness. The presents are wrapped in shiny paper. They're... they're presents. Kids will cry. Parents will call the police. We'll be on the news. 'Local Thieves Steal Teddy Bears and Toy Cars.'" Molly's eyes sparkled. "Exactly! The news! Think of the headlines! 'Criminals Target Toddler's Treasure Trove!' It's perfect! The sheer audacity! The irony! We're not just thieves; we're performance artists! We're stealing joy! Well, plastic joy, but still!" Bill pinched the bridge of his nose. "Performance artists? Molly, we're thieves. We steal things. We don't... we don't steal the *concept* of joy. That's... that's just confusing." Molly grinned, a wicked glint in her eye. "Bill, darling, you're overthinking it. This is the easiest score we've ever had. The presents are practically begging to be taken. They're just... sitting there. Waiting. We just need a little... distraction." "A little distraction?" Bill echoed, his voice rising. "Like what? A fire alarm? That'd bring the whole party down. Or... what? You have a plan involving... what? A balloon animal?" Molly's smile widened. "Bill, my love, you underestimate us. We have a plan. A brilliant plan. It involves... stealth. And misdirection. And a very specific... tool." Bill's eyes narrowed. "A tool? Molly, we don't need tools. We have... well, we have each other. And our wits. Which, frankly, are questionable." Molly patted his arm. "Oh, Bill, you're so sweet. But this time, we do need a tool. A very specific tool. One that will allow us to... access the presents without... disturbing the... uh... the festive atmosphere." Bill stared at her, waiting. "And what tool is that, exactly?" Molly leaned in conspiratorially, her voice dropping to a whisper. "A very large, very convincing... piñata. Bill. We become the piñata. We distract them with our... swinging. While you... you do the actual stealing. From inside the... piñata. It's foolproof!" Bill stared at her, his mouth slightly agape. "You want us to... dress up as a piñata? And swing around? While I... climb into the... piñata... and steal the presents?" Molly nodded vigorously. "Exactly! It's perfect! We're the entertainment! The kids will love it! 'Look, Mommy! The piñata is dancing!' Meanwhile, you're... you're... liberating the loot! It's genius!" Bill slowly shook his head, a slow, disbelieving smile spreading across his face. "Molly... you are absolutely certifiably insane. But... it just might work. The sheer, ridiculous audacity of it. Okay. We do it. We become the piñata. You distract them with your... swinging. I... infiltrate the... piñata. We steal the presents. We leave. We're heroes. Or... we're arrested. Either way, it'll be a story." Molly threw her arms around him, nearly knocking him off balance. "Perfect! Bill, you're a genius! Now, about the costume... I was thinking... something... sparkly. And maybe... feathers? To really sell the 'piñata' vibe?" Bill groaned, but a genuine chuckle escaped him. "Sparkly feathers, Molly. Sparkly feathers. Let's go make our... performance debut." He started walking towards the party, already imagining the chaos to come. "Just... try not to hit anyone with those feathers, okay?"
created_at
"2026-02-17T09:31:07.936729+00:00"
2/17/2026, 9:31:07 AM
model
"arcee-ai/trinity-mini-20251201"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
64
moderation_latency
(null)
generation_time
7206
tokens_prompt
77
tokens_completion
1839
native_tokens_prompt
70
native_tokens_completion
1583
native_tokens_completion_images
(null)
native_tokens_reasoning
591
native_tokens_cached
0
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
usage
0.0002406
router
(null)
provider_responses
0
id
"a2b8954634254d8cbee9f2bbf907d6df"
status
200
is_byok
false
latency
64
endpoint_id
"73d82c49-90ae-4223-a860-83ed0810821f"
provider_name
"Clarifai"
model_permaslug
"arcee-ai/trinity-mini-20251201"
api_type
"completions"
id
"gen-1771320660-GrNQ0U1HJ6Ty33Wsr7HQ"
upstream_id
"a2b8954634254d8cbee9f2bbf907d6df"
total_cost
0.0002406
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Clarifai"
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
767
0.25%
Dialogue to Total Word Ratio
Ratio: 80.96%, Deviation: 10.96%
neededClean
false
wordsTotal
767
wordsDialogue
621
0.1225%