NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYD6P0ACHM4T5138YZBGCM8W
Run Details
Model:
MoonshotAI: Kimi K3 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
48.0s
Cost:
$0.018869
Provider:
Moonshot AI
Evaluation summary
Score
Evaluator
Details Link
0.08%
Matches word count
(details)
36.71%
Dialogue to Total Word Ratio
(details)
18.3965%
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
“Tell me again,” Bill said, “why our big score is a six-year-old’s birthday party.” Molly spread a crayon drawing across the steering wheel. “Because Tommy Turnbull’s grandma is rich, rich people panic-buy, and panic-buying means thirty presents shaped like robots, ponies, and guilt.” “That is a juice box stain.” “That is the south fence.” “It smells like strawberry.” “So will freedom, Bill.” Bill peered through binoculars at the backyard, where a bouncy castle hiccupped children into the air. “I count twelve guests, one magician, a dad grilling with the confidence of a man who has never met smoke, and a piñata shaped like a dragon.” “Mint-condition dragon,” Molly whispered. “Do not emotionally bond with the dragon.” “I’m already naming him Kevin.” “Unname Kevin.” A clown horn honked. Bill ducked. “Was that us?” “No, that’s Gerald, our cover. You’re Gerald’s gloomy assistant.” “I refuse to be gloomy.” “You were born gloomy. You apologize to doors.” Bill adjusted the polka-dot cravat she had stolen from a theater bin. “Fine. But no scaring kids, no grabbing cake from actual hands, and if anyone cries, we become nice immediately.” “Deal. We glide in as entertainment, admire the gift table, create mild confusion with balloon animals—” “You can’t make balloon animals.” “I can make balloon regrets.” Molly produced three limp balloons. “This one’s a snake if you’re brave.” “It looks like a sad sock.” “Perfect. Distraction achieved.” Bill frowned at the crayon map. “Why is the getaway route labeled ‘past the tiny shoes’?” “Because shoes mean adults, adults mean questions, questions mean me saying, ‘We’re from the Department of Surprise Inspection,’ and you sweating through your wig.” “I don’t sweat.” “You once fainted at a raffle.” “It was suspenseful.” Molly checked her watch. “Cake at two, presents at two-thirty, sugar riot at two-forty. We move during the riot.” “And if Tommy catches us?” “We say we’re rehearsing generosity.” “That’s not a thing.” “It will be when I curtsy.” Bill sighed so hard the cravat wilted. “Molly, these are children. Their entire net worth is stickers and one suspicious tooth.” “I know,” she said, softer. “That’s why we only steal the fancy duplicates. Rich grandma always buys two of everything.” “You made that up.” “I made it up confidently.” The back gate opened. A mother waved at them. “Are you the puppet people?” Molly lit up like a stolen chandelier. “Absolutely.” Bill muttered, “We are absolutely not.” “Wonderful! Tommy’s been waiting all year.” All year. Bill looked at the dragon piñata. Kevin looked back with paper eyes. Molly swallowed. “New plan.” “Thank goodness.” “We steal nothing, perform brilliantly, charge triple.” Bill blinked. “That’s just being entertainers.” “Crime’s ugliest cousin.” She grabbed the balloons. “Try to look less like a man who apologizes to doors.” “I make no promises,” Bill said, and followed her toward the tiny shoes.
created_at
"2026-07-25T17:55:31.794Z"
7/25/2026, 5:55:31 PM
model
"moonshotai/kimi-k3-20260715"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2398
moderation_latency
(null)
generation_time
47935
tokens_prompt
75
tokens_completion
1470
native_tokens_prompt
147
native_tokens_completion
1255
native_tokens_completion_images
(null)
native_tokens_reasoning
511
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
0.0188691
router
(null)
provider_responses
0
endpoint_id
"650c9b12-7b6d-444d-8d92-df8daef84d7a"
id
"chatcmpl-6a64f882967b8ab62c55e114"
is_byok
false
latency
2398
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-1785002131-xu4clS0GrtPSpIHDiDhb"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1785002131-Gqzqa4iD0y2KIDxmfS2a"
upstream_id
"chatcmpl-6a64f882967b8ab62c55e114"
total_cost
0.0188691
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
0.08%
Matches word count
n/a
neededClean
false
words
471
36.71%
Dialogue to Total Word Ratio
Ratio: 74.48%, Deviation: 4.48%
neededClean
false
wordsTotal
478
wordsDialogue
356
18.3965%