NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M36SRN7WSSJ12J8GBAJDRZ26
Run Details
Model:
Xiaomi MIMO v2.6 Pro
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
31.9s
Cost:
$0.000805
Provider:
Xiaomi
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0000%
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 a napkin across the diner table and drew a lopsided rectangle with a mustard bottle. "That's the Henderson house," she said. Bill squinted at the napkin. "Why does it look like a potato?" "Because I'm using a condiment, Bill. Focus." "Right. Potato-shaped house. Got it. What's the play?" "Saturday. Little Timmy Henderson turns seven. His parents are throwing a party at two in the afternoon. Approximately twenty-two children in attendance. Grandparents, aunts, uncles. And every single one of those kids is bringing a present." "How do we know this?" "Because it's a birthday party and that's how birthday parties work. I checked the invite. It said clearly — 'No need to bring a gift, your presence is enough.' Which means everyone is absolutely bringing a gift. Guilty parents go big." Bill leaned back and crossed his arms. "So we're robbing children's presents." "We are liberating consumer goods from a poorly secured residential dwelling during a window of distracted supervision. The children are incidental." "They don't sound incidental." "They're kids, Bill. They don't even know what half of these presents are. Do you think a six-year-old needs a steam mop?" "Who's bringing a steam mop to a kid's party?" "Uncle Gerald. He brings household appliances to every event. Last Christmas he gave my nephew a rice cooker." "Okay. Walk me through the entry." "The party starts at two. By two-fifteen, those kids are hopped up on juice boxes and sheet cake. By two-thirty, at least four adults are deep in conversation by the snack table. By two-forty-five, the presents are clustered under a card table in the living room, completely unguarded, because the birthday boy is cutting the cake or opening a piñata." "What if they open the presents during the party?" "They open the presents during the party. That's perfect. The gift bags and wrapping paper get torn apart and shoved aside. The actual gifts get scattered across the floor like a toy store exploded. We don't even have to unwrap anything. We just collect what's already exposed." "And nobody notices two adults stealing from a living room floor?" "That's why we dress as clowns." Bill stared at her. "Come again?" "Children's party. Clown enters. Clown leaves. Nobody questions a clown. It's camouflage." "I don't do clowns." "You don't do clowns, or you don't do the costume?" "The costume itches." "You're robbing a seven-year-old's birthday party and your concern is wardrobe comfort?" "It's a legitimate concern." "Fine. You dress as the dinosaur entertainer. There's no dinosaur entertainer scheduled, which is exactly why no one will question it. A man in a T-Rex suit wanders in, scoops up whatever's nearest the door, and waddles back to the van." "The van." "The van with the window that says 'Candy Makes You Taller!' on the side." "Molly, that van does not say that." "It does now. I made the sticker yesterday." Bill pinched the bridge of his nose. "What's my cut?" "Thirty-five percent." "I'm wearing a dinosaur suit." "Forty." "Split it evenly." "You want fifty percent to waddle around a living room?" "I want fifty percent to waddle around a living room in a heatstroke costume in the middle of the afternoon." Molly studied the napkin drawing. The mustard potato was beginning to dissolve into the paper. "Fine. Fifty-fifty. But if there's anything educational in there, I get first pick." "What's wrong with educational?" "Bill, we are not stealing and then reading. We are stealing and then selling." "Sold. Literally." "Shut up and start stretching. T-Rexes have very limited range of motion and I need you limber."
created_at
"2026-09-23T09:32:49.795Z"
9/23/2026, 9:32:49 AM
model
"xiaomi/mimo-v2.6-pro-20260921"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
772
moderation_latency
(null)
generation_time
31827
tokens_prompt
75
tokens_completion
1004
native_tokens_prompt
71
native_tokens_completion
890
native_tokens_completion_images
(null)
native_tokens_reasoning
64
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
0.000805185
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790155969-mAAfPrmBy20bNbcJLGiY"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790155969-4k4bfNDLVSJyoi1ySc18"
upstream_id
"b2d28838-0169-495b-adb4-ec08b0ad0192_ba868563d7814473a0d5d09124aec723"
provider_responses
0
endpoint_id
"2605413e-4a82-494f-ba65-d75f37ac3f11"
is_byok
false
latency
84
model_permaslug
"xiaomi/mimo-v2.6-pro-20260921"
provider_name
"DeepInfra"
status
429
1
endpoint_id
"4354fc36-b8ab-4ced-aa8c-0efb6b4523f1"
id
"b2d28838-0169-495b-adb4-ec08b0ad0192_ba868563d7814473a0d5d09124aec723"
is_byok
false
latency
697
model_permaslug
"xiaomi/mimo-v2.6-pro-20260921"
provider_name
"Xiaomi"
status
200
total_cost
0.000805185
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
0.00%
Matches word count
n/a
neededClean
false
words
595
0.00%
Dialogue to Total Word Ratio
Ratio: 90.61%, Deviation: 20.61%
neededClean
false
wordsTotal
607
wordsDialogue
550
0.0000%