NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M36T1TR7K20FQWXNN5EB8SN0
Run Details
Model:
Xiaomi MIMO v2.6 Pro
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
4m 13s
Cost:
$0.011892
Provider:
Xiaomi
Evaluation summary
Score
Evaluator
Details Link
97.63%
Matches word count
(details)
99.97%
Dialogue to Total Word Ratio
(details)
98.7985%
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
Bill spread the floor plan across the hood of the car, pinning it down with half-eaten tacos and a novelty lighter shaped like a thumbs-up. The plan was scrawled in red marker on the back of a pizza receipt and included illustrations of children that looked like potatoes with arms. Molly had long since stopped asking where Bill's schematics came from. She turned the receipt over. The other side had a quote from Ecclesiastes. "This is insane," she said, squinting at the drawing. "It's precise. Look — entrance, jungle gym, inflatable slide, and here, the altar of gifts behind the balloon arch. Forty wrapped packages minimum. I spotted a kid carrying in what might be an iPad." Molly set down the receipt and leaned against the car, crossing her arms. The evening air smelled like nearby barbecue and quiet desperation. What Bill was describing was not a bank vault or a museum. It was a seven-year-old's birthday party on Maple Lane. He had cased the location for three days under the guise of being a father named Craig, noting delivery times for the bouncy castle rental and cataloguing the gifts as they arrived. She had to admit the ambition was remarkable. "And what about security? Walk me through the defenses." "Minimal. One clown named Wiggles, possibly one parent supervisor. The clown is the real threat." "Clowns are always the real threat." "Precisely why we need these." He produced two foam earplugs from his jacket pocket and held them up like sacred relics. "The squeaky shoes create psychological disorientation. These neutralize the advantage." "Walk me through the whole thing again. I want to hear every step." He did so at length, gesturing wildly enough that a taco slid off the hood. They would arrive dressed as party entertainers. He would carry a tiered SpongeBob cake with extraordinary care, and she would carry oversized helium balloons that would serve as concealment for the stolen loot. They would enter through the front door, place the cake near the gift table, and wait for the piñata. When it cracked open and fifteen children descended on candy like locusts, no one would be watching the presents. They would load the bags and exit through the back slider to the van. "What if the cake topples?" Molly asked. Bill's face went pale. "Don't put that energy into the universe." "It's a valid concern." "Then I absorb the blow. I become the cushion between that cake and gravity. That's what separates professionals from amateurs." He cracked his knuckles one at a time, which took an unusually long time, then moved on to wardrobe — bow ties, clip-on suspenders, matching red sneakers from the thrift store on Fifth. The theme, he explained, was authority mixed with whimsy. "If we get arrested, you're explaining it to my mother." "Tomorrow at noon. Dress code: birthday. And Molly — act natural. We're birthday people now." Somewhere down the street a dog barked. Molly couldn't shake the feeling the universe was trying to warn them both. She picked up a taco anyway.
created_at
"2026-09-23T09:37:50.349Z"
9/23/2026, 9:37:50 AM
model
"xiaomi/mimo-v2.6-pro-20260921"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
769
moderation_latency
(null)
generation_time
252950
tokens_prompt
75
tokens_completion
11425
native_tokens_prompt
71
native_tokens_completion
13634
native_tokens_completion_images
(null)
native_tokens_reasoning
12986
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
0.011892465
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790156270-KsW36MIGEd8nywW9sV4g"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790156270-APOHntfto9Pk3uboRTsi"
upstream_id
"80ece33e-07c1-4704-97e6-fb17f883d0b7_00744f30433c4add923b986bfda11a8e"
provider_responses
0
endpoint_id
"4354fc36-b8ab-4ced-aa8c-0efb6b4523f1"
id
"80ece33e-07c1-4704-97e6-fb17f883d0b7_00744f30433c4add923b986bfda11a8e"
is_byok
false
latency
662
model_permaslug
"xiaomi/mimo-v2.6-pro-20260921"
provider_name
"Xiaomi"
status
200
total_cost
0.011892465
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
97.63%
Matches word count
n/a
neededClean
false
words
507
99.97%
Dialogue to Total Word Ratio
Ratio: 30.08%, Deviation: 0.08%
neededClean
false
wordsTotal
512
wordsDialogue
154
98.7985%