NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYD6KM6VXM1V5SAZRXH3KWJ0
Run Details
Model:
MoonshotAI: Kimi K3 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
33.4s
Cost:
$0.011489
Provider:
Moonshot AI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
50.0000%
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
The diner booth smelled of burnt coffee and poor decisions, which made it Bill's favorite office. He spread a crayon drawing across the table, flattening it against the ketchup stains. "That's the target?" Molly squinted at the wobbly purple rectangle. "That's the Hendersons' house. I got intel from the inside." Bill tapped a green blob. "This is the bouncy castle. We approach through it." "We approach *through* the bouncy castle?" "Nobody suspects the bouncy castle, Molly. That's Heist Rule Number One." "I thought Heist Rule Number One was 'never work with clowns.'" "That's Rule Two now. I updated the list." Bill sipped his coffee gravely. "Saturday. Two o'clock. Little Timmy Henderson turns seven, and every kid in the neighborhood is bringing a present. Do you know what kids get for their seventh birthday these days?" "Sticker packs?" "Tablets, Molly. Gaming consoles. One kid's bringing a drone. A *drone*. At a seven-year-old's birthday party. What's he going to do with a drone? File his taxes?" Molly leaned back, arms crossed. "So let me get this straight. You want to rob a child's birthday party." "I want to *liberate* roughly three thousand dollars in unopened electronics from a suburban backyard. The child's incidental." "Bill, this is a new low, and I was there for the Easter egg job." "The Easter egg job was profitable!" "They were chocolate, Bill. Everything melted in the van." Bill waved this away like a man swatting a fact. "Here's the plan. You go in dressed as a party entertainer. I found you a costume." He reached under the booth and produced a sequined vest and a plastic wand. "I'm not a magician." "You're a *magic assistant*. There's a difference. The magician has skills. You just hold the hat and go 'ooh.'" He demonstrated, a noise somewhere between a seal and a haunted kettle. "And what are you doing while I'm holding the hat?" "Catering." He puffed out his chest. "I carry the presents table to the van. One trip. Clean." "Bill. People will notice the catering guy walking off with forty wrapped presents." "Not if I'm carrying a tray of juice boxes on top. Misdirection, Molly. It's practically magic too. We should get matching vests." Molly pinched the bridge of her nose. "What's the escape plan?" "Piñata." "That's not a plan, that's a candy dispenser." "You break the piñata, every kid in the yard stampedes one direction, and we walk calmly the other way. It's a smokescreen made of children." "You're using children as a smokescreen." "I'm using their *enthusiasm*. Very renewable resource." Molly stared at the crayon map for a long moment. The green blob did, admittedly, look strategically sound. "What's my cut?" "Sixty-forty." "I want the drone." "You'll crash it into a tree." "I want the drone, Bill." "Fine. But you have to do the 'ooh.' Every time I pull a scarf out of something, you 'ooh.'" "Deal." She grabbed the wand. "And I'm keeping the vest." "Everyone says that. Nobody ever returns the vest."
created_at
"2026-07-25T17:54:13.86Z"
7/25/2026, 5:54:13 PM
model
"moonshotai/kimi-k3-20260715"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2373
moderation_latency
(null)
generation_time
32311
tokens_prompt
75
tokens_completion
814
native_tokens_prompt
147
native_tokens_completion
763
native_tokens_completion_images
(null)
native_tokens_reasoning
42
native_tokens_cached
147
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
0.0114891
router
(null)
provider_responses
0
endpoint_id
"650c9b12-7b6d-444d-8d92-df8daef84d7a"
id
"chatcmpl-6a64f834eaa14129192f02bd"
is_byok
false
latency
2373
model_permaslug
"moonshotai/kimi-k3-20260715"
provider_name
"Moonshot AI"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.18.0; linux; x64))"
http_referer
(null)
request_id
"req-1785002054-ggqn7QsCYG1fdLIHnZDV"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1785002054-3aCcnJARChgNU9MUucnR"
upstream_id
"chatcmpl-6a64f834eaa14129192f02bd"
total_cost
0.0114891
cache_discount
0.0003969
upstream_inference_cost
0
provider_name
"Moonshot AI"
response_cache_source_id
(null)
data_region
"global"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
500
0.00%
Dialogue to Total Word Ratio
Ratio: 75.75%, Deviation: 25.75%
neededClean
false
wordsTotal
503
wordsDialogue
381
50.0000%