NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-02-26-01KJCRWWX09MYQPAMT2ZS74GC6
Run Details
Model:
LFM2 24B
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
13.9s
Cost:
$0.000092
Provider:
Together
Evaluation summary
Score
Evaluator
Details Link
3.62%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
1.8117%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 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 moon hung low over the suburban street as Molly adjusted her black mask, its plastic edges crinkling under her fingers. She whispered to Bill, who was crouched behind a hedge, his face half-hidden by a beanie. “Alright, Bill, this is it. The Great Teddy Bear Heist. You ready?” Bill nodded, adjusting his gloves. “Ready as I’ll ever be. But why do we need to steal every single present? What if they’re not just teddy bears?” Molly snorted. “Because the kid with the biggest smile gets the last one, right? And I’m not letting some brat with a megaphone take it. Besides, think about the chaos—kids screaming, parents panicking. It’s art.” Bill chuckled. “Art? You mean like a crime scene? I’d rather paint a mural.” He paused, then added, “Though I do love a good distraction. Maybe we start with the piñata? It’s already half-broken from last year’s chaos.” “Exactly! We’ll blend in as party crashers. You handle the snacks, I’ll handle the gifts. Simple.” Molly pulled a small flashlight from her pocket, its beam cutting through the dark. “But first, the security. The parents are too busy taking selfies to notice us. Easy pickings.” Bill raised an eyebrow. “Easy pickings? You think this is easy? Last time we tried stealing a birthday cake, the dog ate the frosting.” “That dog had a better plan than most humans,” Molly said, grinning. “This time, we’re smarter. We’ll split up. You take the east wing, I’ll hit the west. Meet back at the tree in ten. No talking unless it’s critical.” “Critical,” Bill repeated, mock-serious. “Like when the kid throws a juice box at your face?” “Exactly. Now, what’s the code for the front door?” Molly checked her watch. “Five seconds. Let’s move before the next round of ‘Happy Birthday’ starts.” They slipped into the shadows, the party’s music muffled by the thick night air. Bill whispered, “You sure about this? What if the kid’s mom sees us?” “She won’t,” Molly said, her voice low. “We’re ghosts. Or… whatever ghosts are good at stealing.” Bill smirked. “Ghosts with a taste for candy. I’m in.” They reached the fence, Molly’s hands brushing the chain-link. “Here’s the plan: you climb, I’ll cut the wire. Then we vanish. No regrets.” “Regrets?” Bill echoed. “You mean like the time you accidentally set the neighbor’s lawn on fire trying to light a cigarette?” “That was a *controlled* fire!” she shot back, laughing. “Anyway, let’s go. The kids are getting bored, and I’d hate to miss the moment the first present gets unwrapped.” Bill nodded, already scaling the fence. “Stay sharp. And Molly?” “Yeah?” “Don’t let the dog near the presents.” She rolled her eyes. “Noted. Now hurry up before I change my mind.” They vanished into the night, the party’s laughter echoing behind them, unaware their greatest heist was already in motion.
created_at
"2026-02-26T10:46:15.55Z"
2/26/2026, 10:46:15 AM
model
"liquid/lfm-2-24b-a2b-20260224"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
261
moderation_latency
(null)
generation_time
13870
tokens_prompt
77
tokens_completion
808
native_tokens_prompt
75
native_tokens_completion
747
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)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
usage
0.00009189
router
(null)
provider_responses
0
endpoint_id
"b5f259c3-1e77-40d6-8fa4-161cdf0e4a7c"
id
"oYuew5K-2j9zxn-9d3edb8b45cac760"
is_byok
false
latency
261
model_permaslug
"liquid/lfm-2-24b-a2b-20260224"
provider_name
"Together"
status
200
api_type
"completions"
id
"gen-1772102775-aot8vWCpeOWpaXqqz157"
upstream_id
"oYuew5K-2j9zxn-9d3edb8b45cac760"
total_cost
0.00009189
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Together"
Evaluation details
Result
Evaluator
Details
Meta Data
3.62%
Matches word count
n/a
neededClean
false
words
476
0.00%
Dialogue to Total Word Ratio
Ratio: 69.17%, Deviation: 19.17%
neededClean
false
wordsTotal
480
wordsDialogue
332
1.8117%