NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-06-01-01KT1HD80CFPXKTGVBV0VP8SW7
Run Details
Model:
MiniMax M3
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
5m 46s
Cost:
$0.019616
Provider:
Minimax
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
73.43%
Dialogue to Total Word Ratio
(details)
86.7170%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 50% 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 kitchen smelled of cold pizza and poor decisions. Molly had pinned the blueprint to the refrigerator with a watermelon magnet, because even criminal masterminds needed snacks. The plan was titled "OPERATION: BIRTHDAY BANDIT" in glitter pen, which she considered a tactical advantage. Glitter, she maintained, was disorienting. "Okay Bill, here's the plan. We hit the Henderson place at four o'clock. Thirty kids, two clowns, one exhausted mother." "Two clowns? You're certain about the clowns?" "Positive. One's called Bozo, the other's Giggles McMurderface. Both clearly from Craigslist." "And the dog?" "The Hendersons have a corgi. I have three sausages in my coat." "Just three?" "Bill, I am a professional. I don't skimp on the meat budget." Bill cradled his coffee like a man warming his soul. He had known Molly for seven years, three prison stints, and one ill-advised pottery class. She was, without question, the most brilliant idiot he had ever met. Her schemes always worked. Her getaways never did. "Fair enough. What about the gift table? It'll be chaos." "Here's the clever part, Bill. We don't touch the gift table." "Then how do we steal the loot?" "We wait until the kids open everything. The paper flies everywhere. The parents clean up, throw it in the bin. And that's when we strike." "You want to steal a trash can?" "A trash can full of PRESENTS, Bill. Parents never throw away unopened gifts. They rewrap them next year and pretend they're new. We'll be swimming in lightly used Spider-Man backpacks and half-eaten candy." "So we're stealing future presents." Outside, rain tapped against the window. The Hendersons lived three blocks away, in a house perpetually decorated for some child's birthday. There was always a child having a birthday. Molly suspected the family was breeding them. "We're stealing from the future to fund our present. It's temporal economics." "You're a monster, Molly." "I'm an opportunist. You're driving the getaway car." "I'm not driving. Last time you let me drive, we ended up in a fountain." "That fountain had it coming, Bill. It was judging us." Bill glanced at the clock. Three hours until the heist. He wondered, not for the first time, how he always ended up holding the getaway car keys while Molly held the philosophical high ground. He suspected kismet, or a clerical error at the universe's main office. "Where are we meeting the fence?" "The usual spot. Behind the inflatable bouncy castle." "There's a bouncy castle at the fence's place?" "There's a bouncy castle at every fence's place, Bill. It's in the bylaws." Molly tapped the blueprint with a sparkly fingernail. The kitchen served as her war room, think tank, and Tuesday afternoon. She had packed the getaway bag already, containing three sausages, rubber gloves, a fake mustache, and an inflatable dinosaur costume, because every escape required plausible deniability and at least one prehistoric reptile. The getaway vehicle, a suspicious minivan with a bumper sticker reading 'MY OTHER CAR IS A GETAWAY CAR,' idled outside. Bill had questions.
created_at
"2026-06-01T12:09:18.609Z"
6/1/2026, 12:09:18 PM
model
"minimax/minimax-m3-20260531"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1241
moderation_latency
(null)
generation_time
345162
tokens_prompt
77
tokens_completion
14437
native_tokens_prompt
236
native_tokens_completion
16332
native_tokens_completion_images
(null)
native_tokens_reasoning
13619
native_tokens_cached
222
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
0.01961592
router
(null)
provider_responses
0
endpoint_id
"6851b352-a7e2-45ad-a69e-67ed1208a416"
id
"066ca96eebea31a9c46822078f8d3e10"
is_byok
false
latency
1241
model_permaslug
"minimax/minimax-m3-20260531"
provider_name
"Minimax"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.16.0; linux; x64))"
http_referer
(null)
request_id
"req-1780315758-YEKsXi31EXAHwMmEpR1M"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1780315758-DponGT9HAQJ7sINfWWBE"
upstream_id
"066ca96eebea31a9c46822078f8d3e10"
total_cost
0.01961592
cache_discount
0.00010656
upstream_inference_cost
0
provider_name
"Minimax"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
500
73.43%
Dialogue to Total Word Ratio
Ratio: 52.49%, Deviation: 2.49%
neededClean
false
wordsTotal
503
wordsDialogue
264
86.7170%