NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-10-27-01M5YHH13CWX03WBZBW95JSPWC
Run Details
Model:
Kimi K3.1 Pro (Reasoning, Max)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
5m 12s
Cost:
$0.110952
Provider:
Moonshot AI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
59.93%
Dialogue to Total Word Ratio
(details)
79.9648%
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 spread the stolen invitation across the steering wheel and smoothed it with one gloved hand. Outside, rainbow balloons bobbed above the Hartleys’ garden, where a bouncy castle shuddered as though digesting something large. Bill adjusted his false moustache and studied the crayoned map she had drawn on a napkin. “The presents are here,” Molly said, tapping a square labelled LOOT. “We enter as entertainers, grab the lot, and leave before the cake.” “Why is the bouncy castle marked doom?” Bill asked. “Because I have seen what happens when thirty children discover sugar,” Molly said. “Our disguises are in the boot. You are a magician. I am your lovely assistant, which is unfair, as I planned everything.” Bill lifted the invitation. “It says fancy dress. We could simply walk in.” “We are not simply walking in,” Molly said. “We are professionals. Professionals rappel.” “From where?” Molly pointed at the oak tree beside the garden. A squirrel sat on the lowest branch, eating a cocktail sausage with the serenity of a king. “Bill, you will distract the parents by doing magic. I will empty the present table into your top hat.” “My top hat holds two scarves and a biscuit.” “Then eat the biscuit.” “What about the children?” “Tell them you are making generosity disappear.” Bill considered this. Inside the garden, someone released a swarm of kazooing toddlers from the castle. They circled a woman carrying a jelly, then scattered when a man in a dinosaur suit arrived with a bubble machine. Bill’s moustache slid down his cheek. “I have changed my mind,” he said. “The dinosaur has a longer reach.” “He is not security.” “Everyone is security at a children’s party. Even the grandparents. Especially the grandparents.” Molly peeled his moustache off his chin and stuck it back above his lip. “Fine,” Molly said. “New plan. We go in through the front door.” Bill blinked. “That is your new plan?” “The front door is the last place anyone expects a burglar.” “It is where they expect guests.” “Exactly. We shall be disguised as guests.” “You are wearing a striped sack labelled SWAG.” Molly looked down at her costume. “This is why you do the magic.” Bill sighed, took the top hat from the boot, and produced a bunch of flowers, three doves, and a parking ticket. “Why on earth did I bring three doves to a robbery? And why is there a parking ticket in my hat?” Molly returned the doves, kept the flowers, and tucked the invitation into her sack. Across the road, the Hartleys’ front door opened. A woman wearing a badge that said MUM waved at them. “Are you the entertainers?” she called. “Yes,” said Molly. “No,” said Bill. The woman squinted. Behind her, children began singing with the brute force of a football crowd. Molly straightened her sack. Bill palmed his biscuit. “Remember,” she whispered, “we leave before the cake.” “And if there are goodie bags?” Molly grinned. “Then, my dear Billdini, we expand the operation.”
created_at
"2026-10-27T11:22:01.206Z"
10/27/2026, 11:22:01 AM
model
"moonshotai/kimi-k3.1-pro-20261023"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1999
moderation_latency
(null)
generation_time
312456
tokens_prompt
75
tokens_completion
10428
native_tokens_prompt
146
native_tokens_completion
11066
native_tokens_completion_images
(null)
native_tokens_reasoning
10348
native_tokens_cached
0
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
0.110952
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.22.1; linux; x64))"
http_referer
(null)
request_id
"req-1793100121-38LOmEjGPLZ1Bq0hthLi"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1793100121-OYVxnpGcWETN8QSkcOr0"
upstream_id
"chatcmpl-6ae0895ab5c4e3b7bc3ac2dd"
provider_responses
0
endpoint_id
"a3fafaac-f097-4ef1-86b3-842854dad5c6"
id
"chatcmpl-6ae0895ab5c4e3b7bc3ac2dd"
is_byok
false
latency
1999
model_permaslug
"moonshotai/kimi-k3.1-pro-20261023"
provider_name
"Moonshot AI"
status
200
total_cost
0.110952
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Moonshot AI"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
500
59.93%
Dialogue to Total Word Ratio
Ratio: 53.20%, Deviation: 3.20%
neededClean
false
wordsTotal
500
wordsDialogue
266
79.9648%