NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-06-01-01KT1GRFCDHYNSRAZ3DDNMHKA8
Run Details
Model:
MiniMax M3
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
42.5s
Cost:
$0.000932
Provider:
Minimax
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0000%
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
Molly crouched behind a recycling bin, binoculars glued to her eyes, studying the bounce house across the street. Bill joined her, unwrapping a cheese stick with his teeth. "So," Molly whispered, smacking the binoculars against her palm. "You ready to relive your childhood trauma for profit?" "Absolutely." Bill chewed thoughtfully. "I haven't been to a birthday party since I was nine. Gerald Henderson pushed me into a piñata. I lost a tooth. I've been planning my revenge ever since." "The piñata or the heist?" "The heist. The piñata's at the bottom of a lake." Molly pulled out a crumpled blueprint that looked suspiciously like the back of a Denny's kids menu. "The party starts at two. Twenty-seven kids, three adults, one clown that looks like he's having an allergic reaction." "Perfect. What's the entry point?" "Through the gift table. I do the crying." Bill paused mid-chew. "The crying?" "Yeah. I show up, mascara running, ask if there's a kid named Lily here. There won't be. By the time they realize I'm not a frantic mom, you're already wrist-deep in wrapping paper." "That's either genius or unhinged." "Same thing, Bill. Same thing." She tapped the menu. "Exit strategy is the bouncy castle. You go out the back, I go out the front, we meet at the car." "What car?" Molly frowned. "The car we don't have yet. I figured you'd steal one on the way." "I was hoping you'd say that." Bill grinned. "What's the target?" "Anything rectangular, under five pounds, and wrapped in glitter paper. Lego sets, action figures, that robotic dog I saw on the commercial. Possibly the shrieking hamster ball thing." "Shrieking hamster ball is non-negotiable." "Obviously." Molly lowered the binoculars. "Now. The clown. He worries me. He's got that look." "The allergy look?" "The 'I peaked in community theater and now I make balloon animals for tips' look. Those clowns are unpredictable." "I once got drop-kicked by a clown at a VFW hall." "Did you deserve it?" "Probably." Bill finished his cheese stick and dusted his hands. "I called his makeup runny." "Right. So we avoid the clown. I distract, you grab, we flee via bouncy castle. Simple." "What if a kid sees us?" Molly straightened up, her face deadly serious. "Bill. We are professionals. We hand that kid a juice box and walk away. No juice box left behind." "Staunch policy." "Staunch policy." She offered a fist bump. "For the hamster ball." "For the hamster ball." Their knuckles met. Somewhere across the street, a child shrieked with laughter. Bill's eye twitched. Molly checked her watch. "Two hours. Go steal us a getaway car. Make it something kid-friendly. We're blending in." "A minivan?" "Bill, you're beautiful. Go."
created_at
"2026-06-01T11:57:58.033Z"
6/1/2026, 11:57:58 AM
model
"minimax/minimax-m3-20260531"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1863
moderation_latency
(null)
generation_time
42502
tokens_prompt
77
tokens_completion
895
native_tokens_prompt
236
native_tokens_completion
762
native_tokens_completion_images
(null)
native_tokens_reasoning
165
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
0.00093192
router
(null)
provider_responses
0
endpoint_id
"6851b352-a7e2-45ad-a69e-67ed1208a416"
id
"066ca6c69011112799ab9aee5551ce6c"
is_byok
false
latency
1863
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-1780315078-FSQLV8WRzotYhfKrik2N"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1780315078-2QpkNbf5FvAytKKjawty"
upstream_id
"066ca6c69011112799ab9aee5551ce6c"
total_cost
0.00093192
cache_discount
0.00010656
upstream_inference_cost
0
provider_name
"Minimax"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
450
0.00%
Dialogue to Total Word Ratio
Ratio: 75.44%, Deviation: 25.44%
neededClean
false
wordsTotal
456
wordsDialogue
344
0.0000%