NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M36SC8A389X2DCZXTTS0V2BM
Run Details
Model:
Xiaomi MIMO v2.6 Flash
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
23.8s
Cost:
$0.000215
Provider:
Xiaomi
Evaluation summary
Score
Evaluator
Details Link
2.01%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
1.0058%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 30% 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 van idled across the street from 42 Birchwood Lane, a modest ranch house draped in balloons and crepe paper that fluttered in the April breeze. A hand-painted sign taped to the mailbox read, "LUCAS TURNS 7 — PARTY STARTS AT NOON." Molly adjusted her ski mask in the rearview mirror and turned to Bill, who was hunched over a folded map of the cul-de-sac like a general planning D-Day. "We go in through the back fence," Molly whispered, tapping a red circle she'd drawn around the yard. "The dog on the flyer is a golden retriever named Captain Waffles. No threat. We hit the gift table, load everything into cooler bags, and we're gone before the last slice of cake is eaten." Bill frowned. "And what about the parents? There'll be at least two dozen adults in there. We can't just walk through the kitchen." "We don't walk through the kitchen. We loop around the garden shed, enter through the sliding glass door on the patio, and grab everything between the piano and the bouncy castle. I timed it. Twelve seconds." "Twelve seconds? That's for the gift table?" "That's for everything." Molly slid a printout across the seat. It was a screenshot of the party invitation that Mrs. Delgado had posted on the neighborhood Facebook group, complete with a list of activities: musical chairs, a magician named Sterling the Amazing, and a piñata. Below it was a photo of last year's haul — thirty-seven wrapped boxes for a child named Owen, most of them Star Wars themed. "That's the kind of haul we're looking at," Molly said. "Lucas's mom runs an Etsy store. His dad imports toys from Shenzhen. The presents alone could be worth a grand, easy." Bill studied the printout, then the map, then the house. A stretch limousine with a TV van's antenna sat in the driveway, presumably belonging to the magician. Children in superhero capes were already visible through the living room windows, bouncing and screaming with their arms raised. "What if Captain Waffles has changed since the flyer?" Bill asked. "Then I have pepper spray and you have a Kong toy with peanut butter. Either way, we're covered." Bill nodded slowly, unconvinced but willing. He pulled a pair of silver bells from his pocket and set them on the dashboard — insurance, he called them, though Molly suspected they were superstition. He checked his watch. "Five minutes until the party starts. We go in during the present opening?" "During the chaos. Sterling will be making a rabbit appear, half the adults will be holding drinks, and all eyes will be on the birthday boy as he tears through wrapping paper. By the time anyone notices the gift table is empty, we'll be halfway to I-95." Molly started the van and pulled it into the alley behind the house, parking beneath a tangle of overhanging oak branches. They sat in silence for a moment, listening to muffled pop music and the distant shriek of a child celebrating something enormous. Bill exhaled. "You know," he said, "this is the most ridiculous thing we've ever done." Molly smiled beneath her mask. "But it's paying work."
created_at
"2026-09-23T09:26:03.336Z"
9/23/2026, 9:26:03 AM
model
"xiaomi/mimo-v2.6-flash-20260921"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2177
moderation_latency
(null)
generation_time
23795
tokens_prompt
75
tokens_completion
854
native_tokens_prompt
71
native_tokens_completion
763
native_tokens_completion_images
(null)
native_tokens_reasoning
78
native_tokens_cached
64
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"
service_tier
(null)
usage
0.000214799
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790155563-jjtJAIdszf1KmDtGazkI"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790155563-eFcRNE9UnFS6bAg253pl"
upstream_id
"032682a5-287c-42d0-b093-29ac95c2eb5a_7f8b35d4a8ef4587a3b43c83d0ef329b"
provider_responses
0
endpoint_id
"6f7ea660-c5e2-492e-8a39-ec254953b3ae"
id
"032682a5-287c-42d0-b093-29ac95c2eb5a_7f8b35d4a8ef4587a3b43c83d0ef329b"
is_byok
false
latency
2101
model_permaslug
"xiaomi/mimo-v2.6-flash-20260921"
provider_name
"Xiaomi"
status
200
total_cost
0.000214799
cache_discount
0.00000878
upstream_inference_cost
0
provider_name
"Xiaomi"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
2.01%
Matches word count
n/a
neededClean
false
words
525
0.00%
Dialogue to Total Word Ratio
Ratio: 46.69%, Deviation: 16.69%
neededClean
false
wordsTotal
529
wordsDialogue
247
1.0058%