NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-17-01KHNFPN9YW6YGQ9SPQJ8JTMXE
Run Details
Model:
Minimax M2.5
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
3m 17s
Cost:
$0.044091
Provider:
Minimax
Evaluation summary
Score
Evaluator
Details Link
99.98%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
49.9920%
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 basement smelled of stale popcorn and cheap cologne, a repurposed hideout that had seen better days. Molly hunched over a cardboard table, a roll of glittery wrapping paper unrolled like a treasure map. Bill lounged in a rickety chair, a half‑eaten donut in one hand and a blueprint of the community center in the other. A neon sign flickered overhead, casting a pink glow on the wall of stolen party hats. Alright, Bill, the target is the Jenkins kid's birthday party tomorrow afternoon at the community center, and we have thirty minutes before the piñata is whacked. The plan was straightforward in theory: arrive early, pretend to be the entertainment, and slip a remote‑controlled vacuum into the pile of gifts. When the birthday boy ripped open the wrapping, the vacuum would suck up the paper and deposit the loot into a hidden compartment disguised as a bounce‑house inflator. The only uncertainty was the DJ, a nervous twenty‑something who loved spontaneous karaoke and could ruin everything if he decided to serenade the crowd. Molly, you really think a clown disguise will work? Those kids are already hyper, and we don't need a panic when balloon animals start popping. To keep the DJ quiet, Molly suggested a bribe: a gift card to the taco truck down the street. Bill, ever the optimist, added a pack of glow‑sticks, reasoning that children love anything that glows and the sticks would serve as a signal when it was time to activate the vacuum. Exactly, Bill, we pose as the entertainment, hand out glittery balloons, and when the music hits the chorus we swap the presents for a fake cake. The duo spent the next hour rehearsing their roles. Molly practiced balloon animals, twisting a deflated tube into a giraffe that looked more like a lopsided dinosaur. Bill tested the confetti cannon, aiming at a stack of old pizza boxes. Each pop sent a burst of iridescent paper fluttering, and the room smelled like a New Year's Eve celebration. If parents start clapping, we'll be stuck doing the Macarena, so I've brought a confetti cannon loaded with glitter bombs – one pop and the kids will cheer. By the time the sun set, they had fine‑tuned the schedule to the minute. At 2:15 the clown would arrive, at 2:30 the gifts would be placed under the table, at 2:45 the music would shift to a high‑energy pop song, and at 2:52 they would trigger the vacuum while the birthday boy blew out his candles. The only variable left was the weather; a sudden rainstorm could turn the bounce‑house into a soggy sponge, but Bill had already packed a waterproof tarp disguised as a giant inflatable dinosaur. We'll wear disposable tuxedos from the thrift store, then dump them in the donation bin after the job – no fingerprints, no trace, just cheap polyester for charity. Alright, I'm in, but if we get caught I'm blaming you for the terrible plan and the inevitable birthday cake shortage.
created_at
"2026-02-17T09:46:19.65158+00:00"
2/17/2026, 9:46:19 AM
model
"minimax/minimax-m2.5-20260211"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
816
moderation_latency
(null)
generation_time
196434
tokens_prompt
77
tokens_completion
16456
native_tokens_prompt
101
native_tokens_completion
18346
native_tokens_completion_images
(null)
native_tokens_reasoning
17719
native_tokens_cached
0
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
usage
0.044091
router
(null)
provider_responses
0
id
"05e36b26804dd8ea672c17e779f0ee77"
status
200
is_byok
false
latency
816
endpoint_id
"489502e5-7555-4113-a3f3-18b8f71e8bb2"
provider_name
"Minimax"
model_permaslug
"minimax/minimax-m2.5-20260211"
api_type
"completions"
id
"gen-1771321382-YQ5XNyu2KzHYL7OV0Gt0"
upstream_id
"05e36b26804dd8ea672c17e779f0ee77"
total_cost
0.044091
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Minimax"
Evaluation details
Result
Evaluator
Details
Meta Data
99.98%
Matches word count
n/a
neededClean
false
words
498
0.00%
Dialogue to Total Word Ratio
Ratio: 0.00%, Deviation: 30.00%
neededClean
false
wordsTotal
498
wordsDialogue
0
49.9920%