NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M36WDFCZDW6HKSHKR1DSWRFG
Run Details
Model:
Xiaomi MIMO v2.6 Pro
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
26.4s
Cost:
$0.000659
Provider:
Xiaomi
Evaluation summary
Score
Evaluator
Details Link
98.71%
Matches word count
(details)
5.83%
Dialogue to Total Word Ratio
(details)
52.2720%
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 the party invitation across the table like it was a classified document. Bill leaned over her shoulder, squinting at the crayon drawings bordering the edges. "It's a child's birthday party," Bill said flatly. "A six-year-old's birthday party, yes," Molly said. "With approximately thirty-five gift bags on the table at any given time." "But they're presents. For a kid." "They're presents for a kid who already has more Lego than the state of Connecticut. We're doing her a favor." "That is not a defense I would ever use in court." "You're not going to court. You're going to a bouncy castle." Bill pinched the bridge of his nose. "Walk me through this again. From the top. Because I keep hoping it'll sound less insane the second time." "It's elegant in its simplicity. The party starts at two. The presents go on a designated table near the back door — I called the venue, and they told me the layout when I pretended to be a catering coordinator." "You pretended to be a catering coordinator." "I said it very convincingly. I even mentioned napkin folds." Bill stared at her. "And then what?" "The guest of honor does cake at three. That's our window. Everyone clusters around the cake, candles get lit, the child screams some kind of barbaric demand for their wishes to be granted, and in that chaos — we walk out the back door with the gifts." "How many gifts are we talking?" "Thirty-five to forty. Plus one inflatable unicorn I have my eye on personally." "You want to steal an inflatable unicorn." "It ties the whole operation together aesthetically." Bill started pacing. Molly watched him make three tight circles around the kitchen table before he spoke again. "What if someone sees us?" "We're wearing fake moustaches and party hats. No one suspects anyone in a party hat." "That's — actually, that's horribly true. But the presents are for kids. The parents bought those gifts. They spent money on them." "They spent money on things like jigsaw puzzles and bath books. This isn't exactly robbing the Vatican." "We are robbing a six-year-old's birthday party, Molly. Let the Vatican comparison stand where it does." Molly stood and held up the invitation between two fingers. "Do you see this? It says 'Barbie Dream House Extravaganza.' Her parents are renting a private event space and hiring a magician named Presto. This child will never miss the gifts." "Because she's six." "Because she has thirty more friends and another birthday next year. The gift supply is renewable." Bill sat back down. He looked at the party invitation one more time. Then he looked at Molly's determined face and the inflatable unicorn sketch she'd already drawn in her heist notebook. "If we get caught," he said slowly, "I want the record to show I objected." "Noted. Now put on this moustache and tell me if it looks natural." He clipped it on and turned to face her. "You look like a walrus." "Perfect. No one suspects a walrus at a children's party."
created_at
"2026-09-23T10:19:09.093Z"
9/23/2026, 10:19:09 AM
model
"xiaomi/mimo-v2.6-pro-20260921"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1887
moderation_latency
(null)
generation_time
26375
tokens_prompt
75
tokens_completion
839
native_tokens_prompt
71
native_tokens_completion
722
native_tokens_completion_images
(null)
native_tokens_reasoning
70
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
0.000659025
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790158749-Bp3HmXzbybLDeomVMjjp"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790158749-KKq92yYx3hr1vAXBVUHG"
upstream_id
"03627937-8961-4414-93af-03e653f9b7ff_c338630d82e94e12822260ce51b2870b"
provider_responses
0
endpoint_id
"4354fc36-b8ab-4ced-aa8c-0efb6b4523f1"
id
"03627937-8961-4414-93af-03e653f9b7ff_c338630d82e94e12822260ce51b2870b"
is_byok
false
latency
1820
model_permaslug
"xiaomi/mimo-v2.6-pro-20260921"
provider_name
"Xiaomi"
status
200
total_cost
0.000659025
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
98.71%
Matches word count
n/a
neededClean
false
words
506
5.83%
Dialogue to Total Word Ratio
Ratio: 77.54%, Deviation: 7.54%
neededClean
false
wordsTotal
512
wordsDialogue
397
52.2720%