NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-17-01KHNETM6B3CJ005ZYB2XPJGSX
Run Details
Model:
Arcee AI: Trinity Mini
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
8.4s
Cost:
$0.000116
Provider:
Clarifai
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
98.96%
Dialogue to Total Word Ratio
(details)
49.4820%
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
Molly slammedher fist on the table. "Bill, we hit the jackpot. The Jones' backyard bash. Candyland theme. Inflatable castle. All the loot is just sitting there, wrapped in shiny paper!" Bill adjusted his glasses, peering over his half-moon spectacles. "Molly, we're not robbing a bank. We're stealing from toddlers. And inflatable castles are... structurally unsound. Also, the parents." "Structurally unsound? Bill, it's a party! Kids bounce, parents drink punch. We're invisible! We grab the presents, bounce out before the cake hits the floor. Easy as pie." "Easy as pie? Molly, pie doesn't scream when you try to steal it. And what about the security? A three-year-old with a juice box is not a security system." Molly grinned, leaning in conspiratorially. "That's the beauty, Bill. The 'security.' They're all distracted. The clown's juggling, the mom's taking pictures, the dad's trying to look cool. We're just... party crashers with a specific shopping list." Bill sighed, rubbing his temples. "Molly, this is madness. We're professional thieves. We don't... we don't... steal from birthday parties." "Professional thieves who need a payday, Bill. And what's more professional than a clean heist? No alarms, no guards, just... presents. Think of the resale value! Action figures, dolls, maybe some... educational toys." Bill hesitated, then slowly nodded. "Okay. Okay, Molly. But we move fast. No talking to the kids. No touching the cake. And for the love of all that's holy, we don't get caught on the security camera. That clown has a camera phone." Molly clapped her hands together. "Perfect! Operation Candyland, commence in... now. Bill, you handle the presents. I'll... I'll... distract the clown. He looks like he could use a good distraction." She winked. "Let's go make some kids cry."
created_at
"2026-02-17T09:27:52.380904+00:00"
2/17/2026, 9:27:52 AM
model
"arcee-ai/trinity-mini-20251201"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2079
moderation_latency
(null)
generation_time
8405
tokens_prompt
77
tokens_completion
910
native_tokens_prompt
70
native_tokens_completion
750
native_tokens_completion_images
(null)
native_tokens_reasoning
443
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
0.00011565
router
(null)
provider_responses
0
id
"1da6ae20251747a29f002a9ab9124236"
status
200
is_byok
false
latency
2076
endpoint_id
"73d82c49-90ae-4223-a860-83ed0810821f"
provider_name
"Clarifai"
model_permaslug
"arcee-ai/trinity-mini-20251201"
api_type
"completions"
id
"gen-1771320463-3W6eQREtTVAGMX6A4431"
upstream_id
"1da6ae20251747a29f002a9ab9124236"
total_cost
0.00011565
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
288
98.96%
Dialogue to Total Word Ratio
Ratio: 86.81%, Deviation: 3.19%
neededClean
false
wordsTotal
288
wordsDialogue
250
49.4820%