NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-24-01KQ01NNZGQ48Q9YM66B1N2WM9
Run Details
Model:
Xiaomi MIMO v2.5
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
17.0s
Cost:
$0.002649
Provider:
Xiaomi
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
2.36%
Dialogue to Total Word Ratio
(details)
1.1799%
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 sprinkler system was a joke—a single drooping rose-shaped nozzle that dribbled water like a sad fountain. Molly and Bill crouched behind a inflatable bouncy castle shaped like a dragon, its deflated neck sagging beside them. “Alright, Bill, review the plan,” Molly whispered, peering through binoculars at the back patio. “Doors open at three. The targets are in that red wagon, guarded only by Gary.” Bill gulped. “Gary the Golden Retriever. He’s six. Our intel says he’ll be distracted by cake.” “Exactly. Phase one: you create a diversion.” Molly pointed. “You offer to help with the piñata.” “The piñata?” Bill’s voice squeaked. “I’m not good with a stick, Molly. Last time, I took out a garden gnome.” “You don’t hit it. You *observe* it. You stand near it, looking contemplative. That’ll be diversion enough. Gary will come to sniff you. Phase two: I slip through the French doors.” “What’s my signal to exit?” “When you hear the words ‘Happy Birthday, Timmy,’ you drop the piñata stick and walk away. Slowly. You don’t run. Running draws attention.” “What if they ask me to pull the rope?” “You say, ‘I think this is a job for the birthday boy.’ It’s poetic. It deflects.” Molly adjusted her black ski mask, which was comically tight. “Now, the entry. The presents are stacked on a table covered in a checkered cloth. Your job is to get a gift bag, the big one with the dinosaurs on it. It’s empty. You hold it open, I load the boxes. Quietly.” “What if a kid sees me?” “You tell them you’re the present sherpa. It’s a job. Kids respect jobs they don’t understand.” “Okay. Okay.” Bill took a shaky breath. “What’s the getaway?” “We walk. We carry the dinosaur bag and another bag of cake. We look like we were invited. If anyone asks, you’re my cousin from Cleveland.” “I’m from Cleveland!” “Perfect. Then you’re Method acting.” A child’s scream erupted from the patio. The game of musical chairs had reached a fever pitch. Gary, the golden guard dog, was currently chasing a squeaky toy thrown by a toddler. His watchful duty was forgotten. “That’s our window,” Molly said. “Go. Be the piñata guy.” Bill stood, brushed grass off his jeans, and marched toward the plastic stick figure holding the candy-filled donkey. Molly melted into the shadows of the house, her sneakers silent on the grass. Minutes later, Bill was standing next to the piñata, looking so awkwardly serious he might as well have been guarding a state secret. Gary trotted over, sniffed his shoe, and lay down at his feet. Perfect. Molly appeared in the doorway, a dinosaur-print bag slung over her shoulder. She gave a tiny nod. Bill’s heart hammered. The patio doors were wide open. She was in. He heard a faint rustle of wrapping paper. Then, a disaster. Timmy, the birthday boy, pointed a sticky finger. “Mommy, why is that lady in my living room?” All heads turned toward the house. Molly froze, a bright blue box halfway to the bag. She slowly turned, a smile plastered on her face. “I’m… the cake decorator! Checking the indoor temperature for frosting integrity.” Timmy’s mom squinted. “We don’t have a cake decorator.” Bill panicked. He needed a bigger diversion. He looked at the piñata. He looked at Gary. He looked at the crowd of expectant five-year-olds. His brain, fueled by pure terror, had a terrible, brilliant idea. “THE PIÑATA!” he bellowed, grabbing the rope. “IT HOLDS THE REAL TREASURE!” He yanked. The stick figure burst open. But instead of candy, a flood of confetti, rubber ducks, and five hundred brightly colored plastic balls cascaded down, engulfing the patio in a chaotic, bouncy explosion. Screams of delight replaced suspicion. Kids dove into the ball pit of prizes. Gary barked, chasing a rolling rubber duck. In the pandemonium, Molly slipped back through the doors, the dinosaur bag now bulging. She grabbed Bill’s arm. “Walk,” she hissed, handing him a bag of cake. “Now.” They strode down the driveway, just two relatives leaving a party, as behind them, children fought over rubber ducks and Gary tried to eat a plastic ball. “See?” Molly said, adjusting her bag. “Told you. Method acting.” Bill stared straight ahead, his face pale. “I need to move to a new city, Molly. A new life. Maybe become a librarian.” “The cake’s good,” Molly offered, pulling a small cupcake from the bag. “Chocolate.” Bill took it, stared at it, and then took a huge bite. It was, admittedly, delicious. The life of crime was a lot of work for such small, sweet rewards.
created_at
"2026-04-24T15:27:57.945Z"
4/24/2026, 3:27:57 PM
model
"xiaomi/mimo-v2.5-20260422"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
606
moderation_latency
(null)
generation_time
16881
tokens_prompt
77
tokens_completion
1526
native_tokens_prompt
311
native_tokens_completion
1293
native_tokens_completion_images
(null)
native_tokens_reasoning
212
native_tokens_cached
192
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"
usage
0.00264896
router
(null)
provider_responses
0
endpoint_id
"2af4bf94-26fa-42b9-bce0-8b11071e105a"
id
"ead1daa28fbd4dacae2b55732b83d318"
is_byok
false
latency
555
model_permaslug
"xiaomi/mimo-v2.5-20260422"
provider_name
"Xiaomi"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.1; linux; x64))"
http_referer
(null)
request_id
"req-1777044477-1ftBE8ROhn9CQR1TKElu"
session_id
(null)
api_type
"completions"
id
"gen-1777044477-QKDSdQWEwah5VkqOjC0I"
upstream_id
"ead1daa28fbd4dacae2b55732b83d318"
total_cost
0.00264896
cache_discount
0.00006144
upstream_inference_cost
0
provider_name
"Xiaomi"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
769
2.36%
Dialogue to Total Word Ratio
Ratio: 41.34%, Deviation: 8.66%
neededClean
false
wordsTotal
774
wordsDialogue
320
1.1799%