NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M36TWRS7VB5WMBA3BTXXDFP7
Run Details
Model:
Xiaomi MIMO v2.6 Flash
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
27.1s
Cost:
$0.000215
Provider:
Xiaomi
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0004%
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 napkin at the diner had a crayon drawing of a house with a balloon outside it. A purple balloon. Bill pressed it flat with his palm and squinted like it was the launch codes for a nuclear submarine. "So you want to break into a six-year-old's birthday party," Bill said, leaning back in the cracked vinyl booth. His coffee had gone cold an hour ago. He kept sipping it anyway. "It's not break in," Molly corrected, tapping her pencil against the napkin. "It's penetrate the perimeter, locate the haul, and exfiltrate before anyone notices." "That's breaking in." "Call it what you want." Molly drew an arrow from the driveway to the front door. "The Des Moines family are going to have forty gifts in there. Tablets, a mini drone, that stuffed elephant that probably costs ninety dollars, and God knows what else." "You want to steal from children." "I want to steal from the parents who spent four thousand dollars funding the children's toy collection." Molly slid the napkin across the table. "It's a victimless crime." "Tell that to the six-year-old holding a balloon animal." The lunch rush had thinned out. Somewhere behind them a cook was grilling something that smelled like regret. Molly unfolded a second napkin on top of the first and drew a crude map of the neighborhood. She marked the back gate with an X and the kitchen entrance with a circle. "Party starts at three. Present opening happens around four-fifteen, after cake, before the piñata." She tapped the circle. "We go in through the back during piñata time. Every adult will be outside watching kids swing at a cardboard burro. The presents sit unguarded in the living room for at least eight minutes." "Piñata time," Bill repeated. "You know it." "What do we need? Rope? Shovels?" "A large duffel bag, comfortable shoes, and a look that says we belong. Bill, this is a children's party. Security is a golden retriever and a guy named Doug." "Doug sounds formidable." "Doug has a beer belly and a lawn chair." Molly leaned forward. "I've watched him for two weeks. He doesn't leave that chair. During piñata time, Doug is asleep with his hat over his face." Bill chewed his pencil and studied the napkin map. The X for the back gate sat near a garbage can. The circle for the kitchen door sat next to a patio where nobody would be looking. The plan had a kind of beautiful stupidity to it. "What's our exit?" he asked. "Same way we came in. Back gate, down the alley, into the car. It takes forty seconds to fill the duffel, thirty seconds to reach the alley, and I timed the walk to the car at two minutes." "And if someone follows us?" "We're just party guests leaving early." Molly folded both napkins and tucked them into her jacket. "We smile, we wave, we keep walking." Bill stared at her for a long moment. The cook dropped a plate somewhere in the kitchen. A woman at the counter laughed too loudly at something on her phone. "When did you learn to plan a heist around a piñata?" he asked. "Sunday. I was bored."
created_at
"2026-09-23T09:52:33.068Z"
9/23/2026, 9:52:33 AM
model
"xiaomi/mimo-v2.6-flash-20260921"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2374
moderation_latency
(null)
generation_time
27045
tokens_prompt
75
tokens_completion
814
native_tokens_prompt
71
native_tokens_completion
731
native_tokens_completion_images
(null)
native_tokens_reasoning
43
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
0.00021462
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790157153-92ffBkmTjKyIHzJq7U0p"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790157153-Nk8mflNZRhJRHfT4658I"
upstream_id
"ec4529e7-531a-42af-87b6-8c566c09e33e_f4c53b607fad4c688be9ab8a3cfc29fb"
provider_responses
0
endpoint_id
"6f7ea660-c5e2-492e-8a39-ec254953b3ae"
id
"ec4529e7-531a-42af-87b6-8c566c09e33e_f4c53b607fad4c688be9ab8a3cfc29fb"
is_byok
false
latency
2341
model_permaslug
"xiaomi/mimo-v2.6-flash-20260921"
provider_name
"Xiaomi"
status
200
total_cost
0.00021462
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
533
0.00%
Dialogue to Total Word Ratio
Ratio: 55.95%, Deviation: 25.95%
neededClean
false
wordsTotal
538
wordsDialogue
301
0.0004%