NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYD54WVD7GSF0J15ACM9PPDF
Run Details
Model:
MoonshotAI: Kimi K3 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
32.7s
Cost:
$0.011204
Provider:
Moonshot AI
Evaluation summary
Score
Evaluator
Details Link
51.93%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
25.9628%
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 spread the crayon-drawn floor plan across the hood of Bill's car. "Target: the Henderson kid's seventh birthday. Saturday, two o'clock." "We're really doing this?" Bill scratched his chin. "Stealing from a child?" "Stealing from thirty children, technically. Every guest brings a gift. That's thirty presents, unopened, stacked on one table, guarded by a magician named Wondrous Walter who I happen to know is legally blind." "That's awful. What's my cut?" "Half, as always. Now, entry. Front door's watched by the grandma. She doesn't miss a thing. Woman once caught a kid palming a third juice box from across the yard." "So the back?" "The back has a bouncy castle. You bounce, you squeak, kids swarm you like pigeons on a chip." Molly tapped the paper. "We go in as entertainment." "You want to infiltrate a birthday party." "I want to infiltrate it as clowns, Bill." "I'm not wearing makeup." "Fine. Face paint." "That's makeup with worse marketing." Molly rolled her eyes. "Balloon artist, then. You twist animals, I work the gift table." "I can't twist balloon animals." "You can't do anything, that's why I'm the brains. You'll learn. It's a dog, a sword, or a snake. Those are your three options and two of them are the same balloon." "What's the snake?" "A balloon you gave up on." Bill sighed and studied the plan. "Where's the getaway?" "Kitchen window, two-fifteen, while Walter does his grand finale. He makes a rabbit disappear." "Where does the rabbit go?" "Into his jacket, Bill. It's not a real mystery. The kids scream, the parents film it, and nobody watches the gift table for ninety full seconds." "Ninety seconds to carry thirty presents?" "Twenty-two. Timmy Henderson already opened eight at his grandmother's party last weekend. Greedy little double-dipper. That's actually how I got the intel." "You interrogated a six-year-old?" "I traded him fruit snacks. He's very corruptible." Bill folded his arms. "And if a kid catches us?" "Then we're funny clowns doing a bit." "You said I could be a balloon artist." "The concept evolved, Bill. Crime is fluid. Keep up." Molly rolled up the floor plan. "One more thing. The piƱata." "What about it?" "It's full of premium chocolate. The good stuff, imported. Mrs. Henderson showed off at the PTA meeting." Molly's eyes gleamed. "We're not leaving without it." "That's not a present, that's party infrastructure." "Everything is a present if you take it home, Bill. That's the philosophy that built this partnership." "I thought the philosophy was 'never steal from anyone poorer than us.'" "It is. And nobody at that party is poorer than us, because you spent our last score on a jet ski." "It was an investment." "We're landlocked, Bill!" "A man's allowed a dream, Molly." She patted his shoulder. "Dream harder. Practice your balloon dog. Saturday we become legends." "Saturday we become the worst people in Ohio." "Legends," Molly repeated, "are rarely beloved."
created_at
"2026-07-25T17:28:42.611Z"
7/25/2026, 5:28:42 PM
model
"moonshotai/kimi-k3-20260715"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2298
moderation_latency
(null)
generation_time
32621
tokens_prompt
75
tokens_completion
808
native_tokens_prompt
147
native_tokens_completion
744
native_tokens_completion_images
(null)
native_tokens_reasoning
40
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
0.0112041
router
(null)
provider_responses
0
endpoint_id
"650c9b12-7b6d-444d-8d92-df8daef84d7a"
id
"chatcmpl-6a64f23b2b42693ae03bd3cc"
is_byok
false
latency
2297
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-1785000522-9xcPqjeo0eP8VR81Fl54"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1785000522-1VHb68kJTZR6RRqd8BFt"
upstream_id
"chatcmpl-6a64f23b2b42693ae03bd3cc"
total_cost
0.0112041
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
51.93%
Matches word count
n/a
neededClean
false
words
484
0.00%
Dialogue to Total Word Ratio
Ratio: 89.80%, Deviation: 19.80%
neededClean
false
wordsTotal
490
wordsDialogue
440
25.9628%