NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-17-01KHNF2FZ5PDWMC77SVZXE8Z5P
Run Details
Model:
Minimax M2.5
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
9m 25s
Cost:
$0.029099
Provider:
Minimax
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
53.58%
Dialogue to Total Word Ratio
(details)
76.7872%
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 perched on the cracked fire escape, the neon sign across the street flickering like a bad disco ball. Bill leaned against the rusted railing, his eyes scanning the map of the community center's floor plan printed on a napkin. The wind carried the distant squeals of children and the faint thump of a bass from the party inside. They whispered their plan, trying not to laugh at the absurdity of stealing presents from a bunch of five‑year‑olds. Molly: "Okay, Bill, here's the plan. The kids will be busy with the clown at three, so we slip in through the door while parents are distracted by cake." Bill: "Right, I'll grab the goodie bags while you go for the piñata. That's the real treasure, right? Tiny toys, sticky hands, glow sticks." Molly: "Exactly. The piñata hangs from the ceiling in main hall. We need to cut it down silently. I've got wire cutters, but we must sync with music." Bill: "The playlist is kid‑friendly, like 'Baby Shark' on repeat. I can sync the cut with chorus, masking the thwack." Molly: "Sure, the parents hover like buzzards. We need a decoy. Remember the big inflatable dinosaur we nicked from the last birthday?" Bill: "Oh, yeah! We'll inflate it in the hallway and claim it's a gift for the birthday boy. Parents will think it's entertainment and keep distance." Molly: "Perfect. We'll slip presents into the dinosaur’s belly. It's enough for ten boxes; no one will suspect a thing." Bill: "I'll handle the camera in the corner. My laser pointer will blind it for few seconds while we exit." Molly glanced at the street below, where a couple of pigeons fought over a crumbs. Bill checked his watch, the battery icon blinking ominously. The community center's windows glowed with streamers, and the occasional laugh punctuated the distant music. They could hear the birthday child's squeal as the balloon popped, a sound like a tiny firework. The plan felt insane, but the prize was worth the risk, and their bright anticipation today. Molly: "Just make sure you don't zap the kid. We're thieves, not monsters. We want the loot, not a tantrum." Bill: "Agreed. No crying kids, no police. Just a heist and a bag of toys." Molly: "Now, the exit strategy: alley has a dumpster, fence is low. We'll toss the dinosaur over and climb out." Bill: "The van is parked two blocks away. We'll sprint like it's the last slice of pizza." Molly: "Let's go over timing one more time. The clown tricks at 3:05, the cake cut at 3:15, piñata drops at 3:30. We strike at 3:35." Bill: "Got it. I'll be at DJ booth, ready to blast the 'Happy Birthday' remix as a signal." Molly: "I'll be near the table, to be a parent. When I see the dinosaur, I'll give the thumbs up." Bill: "Alright, Molly. To the kids' party, to the loot, to the glory." Molly: "And to not getting caught. Ready?" Bill: "Ready as I'll ever be."
created_at
"2026-02-17T09:41:26.485636+00:00"
2/17/2026, 9:41:26 AM
model
"minimax/minimax-m2.5-20260211"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1706
moderation_latency
(null)
generation_time
564579
tokens_prompt
77
tokens_completion
20911
native_tokens_prompt
101
native_tokens_completion
24224
native_tokens_completion_images
(null)
native_tokens_reasoning
23539
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
0.0290991
router
(null)
provider_responses
0
id
"05e368917c8e2b2536a8a8acef68bd39"
status
200
is_byok
false
latency
1705
endpoint_id
"3afb00c4-6135-4c55-96e9-6befa9679b06"
provider_name
"Minimax"
model_permaslug
"minimax/minimax-m2.5-20260211"
api_type
"completions"
id
"gen-1771320721-69rlQYIuaLsLWRH1pKz9"
upstream_id
"05e368917c8e2b2536a8a8acef68bd39"
total_cost
0.0290991
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Minimax"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
501
53.58%
Dialogue to Total Word Ratio
Ratio: 66.47%, Deviation: 3.53%
neededClean
false
wordsTotal
501
wordsDialogue
333
76.7872%