NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M36VEKCJTKR1GKC5AF07T6WZ
Run Details
Model:
Xiaomi MIMO v2.6 Pro
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
33.1s
Cost:
$0.000666
Provider:
Xiaomi
Evaluation summary
Score
Evaluator
Details Link
86.38%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
43.1902%
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 parking lot behind Frankie's Fun Factory & Banquet Hall smelled like stale cotton candy and broken dreams. Molly pressed her face to the grimy window of their van and watched the stream of partygoers filing through the front entrance—parents hauling gift bags, kids streaked with face paint, one boy already crying before the party had even started. She adjusted her binoculars and took notes on a napkin that previously held a gas station burrito. "Fourteen gift tables," she muttered, scribbling furiously. "One near the entrance, two flanking the cake display, and the rest crammed behind the stage like an afterthought. Amateur layout." Bill sat in the driver's seat with his enormous frame wedged behind the steering wheel like a stuffed bear in a toy claw machine. He was studying the party floor plan they'd obtained through their usual method—Molly had called the venue pretending to be a grandmother planning a retirement party and had charmed the teenage receptionist into sending a detailed layout via text. "How many chaperones we talking?" Bill asked. "Eleven. But eight of them are moms who've already opened the complimentary wine bar, so realistically we're looking at three threats." Bill nodded slowly. "What about the birthday kid? Last time we underestimated the birthday kid." "A six-year-old named Chet. His mother told the receptionist he wants to be a paleontologist. He'll be elbow-deep in a dinosaur cake by 2:15. We strike at 2:17." "Two minutes after cake. Bold." "Two minutes after cake is when every adult in that room is either filming their kid with buttercream on their face or refilling their cup. It's the most psychologically vulnerable moment in any children's party. I read a paper about it." "You read a paper about children's party vulnerabilities." "I read a lot of things, Bill." She tossed the napkin into the back of the van where it landed atop an impressive pile of other napkins, each covered in equally detailed scribbles. They'd been casing Frankie's Fun Factory for three weeks, attending no fewer than four separate children's parties disguised as uncles, cousins, and once as a clown named Professor Squelch. Molly had despised the clown gig but had to admit the oversized shoes were surprisingly good for concealing lockpicks. Bill pulled out a pair of tiny latex gloves meant for a child's hands and stretched them over his own massive fingers. They snapped back into place instantly. "These aren't going to work." "That's the third size up. Your fingers are an anomaly of nature and I refuse to be held responsible for it." "Fine. What about the getaway route?" "Through the bounce house. No adult will follow us into the bounce house. It's sacred ground. They bounce on one side, we bounce on the other, and by the time they realize what's happened, we're out the emergency exit behind the claw machines." Bill stared at her. "The emergency exit behind the claw machines leads to a dumpster." "A dumpster full of presents is still a dumpster full of presents, Bill. Now put on your party hat."
created_at
"2026-09-23T10:02:17.369Z"
9/23/2026, 10:02:17 AM
model
"xiaomi/mimo-v2.6-pro-20260921"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
3053
moderation_latency
(null)
generation_time
33106
tokens_prompt
75
tokens_completion
853
native_tokens_prompt
71
native_tokens_completion
730
native_tokens_completion_images
(null)
native_tokens_reasoning
75
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
0.000665985
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790157737-VrxbcXT7Zb8L7UzHSzKz"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790157737-TiCUI8lPaPT0DvztnQeI"
upstream_id
"25f5e1c6-60a4-46b8-854f-60346922835e_b01fe7adb9654359b59db766894f8b17"
provider_responses
0
endpoint_id
"4354fc36-b8ab-4ced-aa8c-0efb6b4523f1"
id
"25f5e1c6-60a4-46b8-854f-60346922835e_b01fe7adb9654359b59db766894f8b17"
is_byok
false
latency
2955
model_permaslug
"xiaomi/mimo-v2.6-pro-20260921"
provider_name
"Xiaomi"
status
200
total_cost
0.000665985
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
86.38%
Matches word count
n/a
neededClean
false
words
511
0.00%
Dialogue to Total Word Ratio
Ratio: 50.39%, Deviation: 20.39%
neededClean
false
wordsTotal
514
wordsDialogue
259
43.1902%