NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M36VDBNAVEB38DRM39SDQ508
Run Details
Model:
Xiaomi MIMO v2.6 Flash
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
37.7s
Cost:
$0.000222
Provider:
Xiaomi
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
1.42%
Dialogue to Total Word Ratio
(details)
0.7077%
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 leaned against the brick wall behind the venue, adjusting her ski mask as Bill approached with two lukewarm coffees from a paper cup carrier. "So you're telling me you didn't scout the location?" Bill asked, handing her a coffee. "I scouted it. I looked at it on Google Maps. That counts." "That's not scouting, Molly. That's using a computer." "Bill, for someone who robbed a chicken restaurant last summer, you're being awfully judgmental." "I didn't rob it. I found money behind the cash register. There's a difference." "The door code, Molly. What's the door code?" "Two-seven-four-nine." "You made that up." "I did not make it up. I made it up in front of a security camera so someone would think I made it up, and then I used the real one." "That is... actually genius." "Thank you." "But we still don't have a real door code." "I said there were two codes. What is wrong with your ears?" "Fine. What's the real code?" "Eight-eight-two-four." "And you got that from... where?" "A man named Derek. He works there. He has a very open policy about door codes and a deep love of margarita glasses." "So Derek is our inside man?" "Derek doesn't know he's our inside man. But he told me the code because I told him I was a party planner." "You don't look like a party planner." "I look like everything, Bill. That's my gift." "Alright. So we get in with the code. Then what?" "Then we walk through, grab every present off the gift table, and walk out the same door." "The same door we walked in through?" "Yes, Bill. The same door we walked in through. We are not crawling through vents. We are adults." "How many kids will be at this thing?" "Fourteen. All under the age of eight. Two of them can barely walk. One of them is exclusively crawling." "That sounds like it'll be loud." "It will be screaming loud. Which is perfect for us. No one is going to hear us zipping open gift bags." "What about the parents? There'll be parents there." "Molly grabbed Bill by the collar and pulled him closer. "There will be a host. Two parents. One of them is on her phone the entire time because her husband left last Tuesday. The other one is doing a crossword puzzle because he does not care about his son's birthday." "And you know this how?" "Molly and I am very thorough." "That's disturbing." "Are we doing this or not?" Bill paused. He took a sip of coffee and winced. "This is terrible coffee." "It's from a cart behind a gas station. Set expectations accordingly." "Fine. We do it. Gift table grabs. Quick in and out." "Thank you." "Once we have the gifts, we sell them online?" "We sell them to an online auction. I already listed them." "You already listed them? Before we even have them?" "I posted pre-Christmas gift bundles on three different platforms. People are already bidding." "You are a criminal mastermind." "I know." "An unwell, deeply confusing criminal mastermind." "Also yes." Bill finished his coffee and crushed the cup. "Alright, Molly. Let's go steal some presents from a bunch of toddlers." "That's the spirit."
created_at
"2026-09-23T10:01:36.693Z"
9/23/2026, 10:01:36 AM
model
"xiaomi/mimo-v2.6-flash-20260921"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1013
moderation_latency
(null)
generation_time
27570
tokens_prompt
75
tokens_completion
852
native_tokens_prompt
71
native_tokens_completion
758
native_tokens_completion_images
(null)
native_tokens_reasoning
50
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
0.00022218
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790157696-Q7pIjnh9NCP9OsFp8hAv"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790157696-Rr7ggeTd8Dh6MJVEOPIA"
upstream_id
"1036f386-43ef-4a78-9506-cc3977b13e56_f554fa78f7fa40b18d46c58d25da9fb7"
provider_responses
0
endpoint_id
"8f1a1d4b-adc8-465f-860a-4dc2aad88765"
is_byok
false
latency
79
model_permaslug
"xiaomi/mimo-v2.6-flash-20260921"
provider_name
"DeepInfra"
status
504
1
endpoint_id
"6f7ea660-c5e2-492e-8a39-ec254953b3ae"
id
"1036f386-43ef-4a78-9506-cc3977b13e56_f554fa78f7fa40b18d46c58d25da9fb7"
is_byok
false
latency
885
model_permaslug
"xiaomi/mimo-v2.6-flash-20260921"
provider_name
"Xiaomi"
status
200
total_cost
0.00022218
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
538
1.42%
Dialogue to Total Word Ratio
Ratio: 79.23%, Deviation: 9.23%
neededClean
false
wordsTotal
544
wordsDialogue
431
0.7077%