NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-24-01KQ02ZXGR5GZGGPZ6CZDMA7WM
Run Details
Model:
Xiaomi MIMO v2.5 Pro
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
18.8s
Cost:
$0.002732
Provider:
Xiaomi
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
3.84%
Dialogue to Total Word Ratio
(details)
1.9191%
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 spread the blueprint across the kitchen table with the gravity of a surgeon preparing for an operation. Bill leaned over her shoulder, squinting at the crayon-drawn layout. "Okay, walk me through it again," Bill said. "Party starts at two. PiƱata at three. That's our window." Molly tapped a circle labeled 'Presents' with the end of her pencil. "The parents are all distracted with cake duty. The kids are blindfolded and swinging at a donkey filled with candy. Nobody's watching the gift table." "What about the clown?" "What clown?" "There's always a clown, Molly." She sighed and pulled out a photograph. "His name is Doodles. He does balloon animals. I've been watching him for weeks." Bill stared at her. "You've been surveilling a children's party clown for weeks." "Professionalism, Bill." "You followed a man named Doodles." "To the grocery store. To the laundromat. To his mother's house on Thursdays." Molly straightened up proudly. "He's left-handed, lactose intolerant, and he can't make anything more complex than a poodle. He's not a threat." "Okay, so walk me through entry." "We go in through the bounce house." Bill blinked. "The bounce house." "It's inflatable. We deflate the back corner, slip in through the vinyl seam, reinflate, and emerge from inside like we belong there. Nobody questions a man climbing out of a bounce house at a children's party. They assume you're someone's uncle." "What if someone sees us cutting the vinyl?" "That's why I brought this." Molly produced a pizza delivery bag. "We walk in carrying pizza. People see pizza, they don't see faces. It's science." "Where's the pizza?" "There's no pizza. The bag is empty." "So we're carrying an empty pizza bag into a children's party to infiltrate a bounce house in order to steal presents from six-year-olds." "When you say it like that, it sounds ridiculous." "It IS ridiculous, Molly." "Ridiculously brilliant." She circled the gift table on the blueprint. "Here's the beautiful part. Every kid writes their name on their presents. So we know exactly which ones are expensive. Anything with 'Love, Grandma' goes in the bag. Grandmas always spend big." "And the getaway?" "Bicycle." "A bicycle." "Tandem bicycle. I got a great deal on it." "You bought a tandem bicycle." "It was forty dollars, Bill. We'd be losing money if we didn't buy it." Bill rubbed his temples. "What about the presents? We'll have thirty pounds of wrapped gifts. Where do they go on a tandem bicycle?" Molly pointed to the blueprint again. "The basket." "There's a basket?" "There's a basket." Bill exhaled slowly. "Okay. Fine. What's our exit strategy once we have the goods?" "We pedal two blocks to the van." "Now there's a van?" "I always have a van, Bill." He stared at her for a long moment. "What if something goes wrong?" Molly smiled and pulled out a bag of party hats. "We put these on and act like parents." "Those are lavender. No parent wears lavender." "They do at THIS party. Theme is Fairy Garden." She tossed him a pair of glittery wings. "You're wearing these." Bill caught them against his chest and looked down at the sequined fabric. "I want a raise." "We split fifty-fifty. That's the deal." "I want a raise in dignity." Molly patted his shoulder. "Dignity doesn't pay the bills, partner. Now let's go steal some Legos."
created_at
"2026-04-24T15:51:01.923Z"
4/24/2026, 3:51:01 PM
model
"xiaomi/mimo-v2.5-pro-20260422"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
732
moderation_latency
(null)
generation_time
18712
tokens_prompt
77
tokens_completion
1016
native_tokens_prompt
315
native_tokens_completion
874
native_tokens_completion_images
(null)
native_tokens_reasoning
94
native_tokens_cached
256
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
0.0027322
router
(null)
provider_responses
0
endpoint_id
"5c24ecc9-5ba1-4af0-8877-3402889e7d0a"
id
"fa21a2a51180424ab5068d2cf510fa35"
is_byok
false
latency
650
model_permaslug
"xiaomi/mimo-v2.5-pro-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-1777045861-HOTaRDqzNtum6oJwceoO"
session_id
(null)
api_type
"completions"
id
"gen-1777045861-iNUUvVKHddB9R35IFHf9"
upstream_id
"fa21a2a51180424ab5068d2cf510fa35"
total_cost
0.0027322
cache_discount
0.0002048
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
556
3.84%
Dialogue to Total Word Ratio
Ratio: 78.07%, Deviation: 8.07%
neededClean
false
wordsTotal
561
wordsDialogue
438
1.9191%