NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-17-01KHNFMF6YM6D9KTWZ38M757AV
Run Details
Model:
Arcee AI: Trinity Mini
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
5.4s
Cost:
$0.000132
Provider:
Clarifai
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0000%
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
Molly leaned againstthe fence, squinting at the backyard party. "Look at that, Bill. Pure gold. Kids everywhere, parents too busy sipping punch to notice a thing. And the presents! Stacked like a pyramid of loot." Bill adjusted his glasses, peering through his binoculars. "Molly, it's a birthday party. The presents are... well, presents. Toys. Probably not even worth the effort of a crowbar." "Effort? Bill, you're forgetting the value of novelty!" Molly gestured wildly. "Think about it. A brand new scooter! A dollhouse! A giant teddy bear! That's liquid capital right there. Parents will pay anything to get those back. We're not stealing toys; we're securing high-value, low-risk assets." Bill sighed, rubbing his temples. "Molly, the risk is high. Kids cry. Parents get angry. Security guards? There's always some dad with a baseball bat who thinks he's a security consultant. And the presents... they're wrapped in shiny paper. How do we even get them out? A pillowcase? That screams 'thief' louder than a fire alarm." Molly grinned, already scaling the fence. "Problem solved. We use the ladder. Simple. Distract the parents with the cake. Everyone loves cake. While they're oohing and aahing over the frosting, we slip in, grab the presents, and vanish before anyone realizes the piñata's been... relocated." Bill followed, muttering. "Relocated? Molly, that's not a word. And cake? You think parents will be distracted by cake? They're already distracted by the kids. This is madness." "Madness is my middle name," Molly declared, reaching the ladder. "Now, climb. We've got a heist to pull off. And remember, Bill, in the world of high-stakes toy retrieval, there are no small jobs." Bill climbed, his movements slow and deliberate. "High-stakes toy retrieval? Molly, we're stealing from a kindergarten. The stakes are... low. The retrieval part is purely hypothetical." Molly ignored him, hoisting the ladder against the fence. "Focus, Bill. Focus on the plan. Distract with cake. Grab presents. Escape. Profit. Simple." Bill peered over the fence at the laughing children and the colorful decorations. "Profit? Molly, the parents will just buy new presents. We're not stealing from a museum. This is... this is stealing from a birthday. It's like robbing a piñata that's already been smashed." Molly gave him a look. "Exactly. The piñata's been smashed. The loot is loose. Now, climb down. We've got a party to crash."
created_at
"2026-02-17T09:41:56.466422+00:00"
2/17/2026, 9:41:56 AM
model
"arcee-ai/trinity-mini-20251201"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
60
moderation_latency
(null)
generation_time
5336
tokens_prompt
77
tokens_completion
1004
native_tokens_prompt
70
native_tokens_completion
857
native_tokens_completion_images
(null)
native_tokens_reasoning
374
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
0.0001317
router
(null)
provider_responses
0
id
"7b8814157090441abe5d85649bc25680"
status
200
is_byok
false
latency
60
endpoint_id
"73d82c49-90ae-4223-a860-83ed0810821f"
provider_name
"Clarifai"
model_permaslug
"arcee-ai/trinity-mini-20251201"
api_type
"completions"
id
"gen-1771321310-5bqakOkTMjEYzsDwJXEd"
upstream_id
"7b8814157090441abe5d85649bc25680"
total_cost
0.0001317
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
393
0.00%
Dialogue to Total Word Ratio
Ratio: 81.42%, Deviation: 31.42%
neededClean
false
wordsTotal
393
wordsDialogue
320
0.0000%