NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M36SPC7SXYFDRR8ENPKPVMVX
Run Details
Model:
Xiaomi MIMO v2.6 Pro
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
32.0s
Cost:
$0.000728
Provider:
Xiaomi
Evaluation summary
Score
Evaluator
Details Link
1.04%
Matches word count
(details)
0.04%
Dialogue to Total Word Ratio
(details)
0.5386%
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
Bill unfolded the laminated blueprint across the hood of the car and tapped his finger on the giant inflatable bouncy castle. "This is where they'll hide the presents," he said. Molly squinted at the blueprint. "That's a diagram of the moon." "It's a diagram of the backyard, Molly." "The backyard has craters?" "That's the sandbox. Look, the point is, the presents will be stacked on the picnic table, right here, behind the obstacle course of seven-year-olds." Molly leaned against the car and crossed her arms. "Seven-year-olds. You sure you can handle that? Last week you got tripped up by a cat." "The cat was *lying in wait*." "It was asleep." "It was a tactical nap. Very professional. Now listen, we hit during the cake. That's when every parent and kid is focused on the candles. Some kid named Brayden is turning eight. Brayden." "I hate him already." "Everyone hates Brayden. I hate Brayden and I've never met him. But the hatred is key. It'll fuel us through the extraction." "What exactly is the extraction?" "We load the presents into the pony." Molly stared at him. "The pony." "Brayden's parents hired a pony. It's in the diagram." "That's a—fine. What kind of security are we looking at?" "Minimal. One balloon artist, potentially two clowns." "Two clowns?" "One confirmed, one rumored. The rumor came from the piñata guy, who I trust implicitly." "You talked to the piñata guy?" "I brought him coffee. He's chatty. Anyway, the clowns are the real threat. They're unpredictable. Unprofessional. No code of ethics." "Worse than the cat?" "Different kind of threat. The cat was physical. Clowns are psychological. You'll need mental preparation." "I'll manage." "Okay. So the signal for go-time is when they sing 'Happy Birthday.' Second verse—there's only one verse, but I'm counting the clapping at the end as a second verse. When the clapping starts, we move." "What if they clap irregularly?" "Then we improvise. We're professionals." "You stole a fire truck last month and drove it to a Denny's." "Into a Denny's. Through the window. But that was off the books. Tonight, we execute clean. I even have costumes." Bill reached into the back seat and pulled out two oversized animal onesies. "We go in as mascots. I'm the rabbit. You're the—" "A fox. Of course. The fox steals the presents. Real subtle." "The kids won't suspect a fox. They love foxes. There's a whole show about it." Molly took the fox costume and held it up. It smelled like animal crackers and defeat. "I have three questions." "Go ahead." "One: how do we get the presents out? Two: is there a backup plan? Three: are we really doing this?" Bill cracked his knuckles. "The presents go out through the side fence in the trash bags from the birthday boy's gift table. The backup plan is chaos—if things go sideways, I release the piñata early and everything descends into madness. And three—" He put on the rabbit head. His voice became muffled and strange. "Yes, Molly. We are absolutely doing this." From inside the rabbit suit, he added: "Put on the fox. We've got a birthday to ruin."
created_at
"2026-09-23T09:31:35.039Z"
9/23/2026, 9:31:35 AM
model
"xiaomi/mimo-v2.6-pro-20260921"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2384
moderation_latency
(null)
generation_time
32028
tokens_prompt
75
tokens_completion
873
native_tokens_prompt
71
native_tokens_completion
801
native_tokens_completion_images
(null)
native_tokens_reasoning
65
native_tokens_cached
0
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
0.000727755
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790155895-DhYitKWYiQIY0PYhskR5"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790155895-63qqdxI6p8H9apUKEmmp"
upstream_id
"3149bafb-196a-4a72-bd04-0cdd823a7f44_1c0880711b5c40ceb96ba40278da6039"
provider_responses
0
endpoint_id
"4354fc36-b8ab-4ced-aa8c-0efb6b4523f1"
id
"3149bafb-196a-4a72-bd04-0cdd823a7f44_1c0880711b5c40ceb96ba40278da6039"
is_byok
false
latency
2306
model_permaslug
"xiaomi/mimo-v2.6-pro-20260921"
provider_name
"Xiaomi"
status
200
total_cost
0.000727755
cache_discount
(null)
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
1.04%
Matches word count
n/a
neededClean
false
words
526
0.04%
Dialogue to Total Word Ratio
Ratio: 82.49%, Deviation: 12.49%
neededClean
false
wordsTotal
531
wordsDialogue
438
0.5386%