NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M36TVQBA7J0E98PEANAPG62K
Run Details
Model:
Xiaomi MIMO v2.6 Flash
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
1m 18s
Cost:
$0.000212
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
99.38%
Matches word count
(details)
0.02%
Dialogue to Total Word Ratio
(details)
49.7005%
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
Bill pressed himself against the hedge and peered through the gap in the fence. "Are you sure about this, Molly?" "Sure as sure gets, Bill. Look at that list — thirty kids, forty-two presents. Do you know what forty-two presents is worth?" "That's not the kind of arithmetic I like when it involves toddlers." "It's called retirement, Bill. Early retirement. Before we're seventy." Bill wiped his palms on his trousers. "Okay, but walk me through it again. The whole plan." "Right. You're going to go in the front door wearing the clown suit." "I hate the clown suit." "Everyone hates the clown suit. That's why it works. You distract the adults while I come in through the sliding glass door in the backyard and grab everything under that big table with the streamers." "What if the kids follow me?" "The kids won't follow you, Bill. You're a clown. They've been conditioned to trust clowns. It's practically in their DNA." "That's not reassuring in the slightest, Molly." "It doesn't need to be reassuring. It needs to be effective. There's a difference." Bill sighed. "What about the wrapping paper? Do we unwrap things here or take them as they are?" "Take them as they are. What kind of amateur steals presents unwrapped? We grab, we go, we sell. Clean work." "Where do we sell stolen presents?" "I know a guy." "You always know a guy." "And it always works, doesn't it?" Bill rubbed the back of his neck. "I suppose. But what if someone recognizes us?" "Who's going to recognize us, Bill? We're wearing clown suits and birthday party hats. We'll look like we belong." "We'll look like we belong at exactly one thing, and it's a crime scene." Molly grinned. "Exactly. So what's the entry point again?" "You said sliding glass door. Backyard." "And your job?" "Front door. Clown suit. Make balloon animals, sob when the kids aren't looking, get everyone looking at me." "You've got it. One more thing — the get-away car." "The hatchback?" "No, Bill. The get-away car. Park it around the corner on Maple Street. Engine running. Warm." "How am I supposed to park around the corner with the engine running? Who sits in a car with the engine running on Maple Street?" "You'll be in the car, you dolt. I'll signal you when I've got the bag." "Fine. Fine. But if I get arrested in this clown suit, I swear, Molly, I'm telling them everything." "You won't get arrested. You're a clown at a birthday party. Worst thing that happens is someone asks you to do a balloon sword." Bill adjusted the ridiculous ruffled collar around his neck. "I actually do make a pretty good balloon sword." "I know you do, sweetheart. That's why I picked you." Molly tapped the fence once, twice, and slipped into the darkness beside the hedge. Bill straightened up, shoved a rubber nose onto his face, and took a breath. Thirty seconds later, a car door clicked shut, and a horn honked — twice, then once. It was time.
created_at
"2026-09-23T09:51:58.832Z"
9/23/2026, 9:51:58 AM
model
"xiaomi/mimo-v2.6-flash-20260921"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
295
moderation_latency
(null)
generation_time
77718
tokens_prompt
75
tokens_completion
801
native_tokens_prompt
71
native_tokens_completion
720
native_tokens_completion_images
(null)
native_tokens_reasoning
44
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
0.00021154
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790157118-EwT5bUQu8acQJa37M23w"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790157118-cBjh4ekePylmWZ0XeyJF"
upstream_id
"chatcmpl-RDKnbpgl6b97H19qjH43dvM0"
provider_responses
0
endpoint_id
"8f1a1d4b-adc8-465f-860a-4dc2aad88765"
id
"chatcmpl-RDKnbpgl6b97H19qjH43dvM0"
is_byok
false
latency
80
model_permaslug
"xiaomi/mimo-v2.6-flash-20260921"
provider_name
"DeepInfra"
status
200
total_cost
0.00021154
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepInfra"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
99.38%
Matches word count
n/a
neededClean
false
words
505
0.02%
Dialogue to Total Word Ratio
Ratio: 82.91%, Deviation: 12.91%
neededClean
false
wordsTotal
509
wordsDialogue
422
49.7005%