NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYD4VB9C9R2YQAQTSFRRDK1P
Run Details
Model:
MoonshotAI: Kimi K3 (Reasoning, Low)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
29.5s
Cost:
$0.010934
Provider:
Moonshot AI
Evaluation summary
Score
Evaluator
Details Link
86.38%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
43.1902%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 30% 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 van smelled like old fries and worse decisions, which was fitting, Molly thought, because they were parked outside a bouncy castle at eleven in the morning on a Saturday. "Remind me again why we're robbing a six-year-old?" Bill asked, adjusting the clown wig that itched something fierce. "Because, William, little Timmy Henderson's birthday party has forty-three confirmed presents, and his father collects rare coins and wraps them as gag gifts." Molly smoothed her own wig, a fetching neon pink number. "One of those 'gag gifts' is an 1804 silver dollar worth more than this van. More than a hundred of these vans." "And the bouncy castle?" "Distraction. You get stuck in it." "I get what?" "You're the distraction, Bill. You're six-foot-three. The moment you try to bounce, every parent in that yard will be filming you instead of watching the gift table." Bill stared out the windshield at the inflatable monstrosity, which was shaped like a dragon and appeared to be judging him. "I don't like this plan. The last plan with me as the distraction ended with me in a koi pond." "The koi pond was your fault. You were told to walk, not run." "There was a swan, Molly." Molly pulled a duffel bag from under the seat and began stuffing it with empty wrapping paper. "Swap technique. We replace the presents with pre-wrapped boxes of identical size and weight. Nobody notices until the unwrapping, and by then we're three counties away eating pancakes." "What's in the fake boxes?" "Rocks. Very festive rocks." "You're a monster. I love working with you." They climbed out of the van, which had "PARTY PALS ENTERTAINMENT" stenciled on the side in letters that were already peeling. Bill had done the stenciling himself, badly, and Molly had decided it added authenticity. Real party entertainers never looked quite legitimate. "Balloon animal skills?" she asked as they approached the gate. "I can make a snake." "That's just an un-inflated balloon." "I can make a worm." "Also just an un-inflated balloon." "Then the children will marvel at my range." A woman in a sun hat intercepted them at the fence, clipboard in hand, smelling of sunscreen and suspicion. "You're the entertainment? You're forty minutes late." "Traffic," Molly said smoothly. "Clown convention. You wouldn't believe the congestion." The woman squinted at Bill, who waved and honked his nose. The honk came out wrong, more of a wheeze, and a nearby toddler began to cry. "He's method," Molly added. Inside the yard, the gift table groaned under a mountain of shiny packages, and Molly's eyes went soft the way they only did around other people's valuables. She clocked the exits, counted the parents, and located the coin-sized box, third from the left, wrapped in dinosaur paper. "Bill," she murmured. "The dragon awaits." Bill sighed the sigh of a man who had made many poor choices and was about to make one more. "If I die in that castle, tell my story honestly." "I'll say you bounced bravely." And with that, the world's least qualified clown waddled toward destiny.
created_at
"2026-07-25T17:23:29.715Z"
7/25/2026, 5:23:29 PM
model
"moonshotai/kimi-k3-20260715"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2545
moderation_latency
(null)
generation_time
29403
tokens_prompt
75
tokens_completion
800
native_tokens_prompt
147
native_tokens_completion
726
native_tokens_completion_images
(null)
native_tokens_reasoning
17
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
0.0109341
router
(null)
provider_responses
0
endpoint_id
"650c9b12-7b6d-444d-8d92-df8daef84d7a"
id
"chatcmpl-6a64f11376dcf69d31951e16"
is_byok
false
latency
2545
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-1785000209-8CyA1QjeTFQ4s8GuVQjc"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1785000209-E1m7hk67AAwKkQPPzrQE"
upstream_id
"chatcmpl-6a64f11376dcf69d31951e16"
total_cost
0.0109341
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
86.38%
Matches word count
n/a
neededClean
false
words
511
0.00%
Dialogue to Total Word Ratio
Ratio: 48.55%, Deviation: 18.55%
neededClean
false
wordsTotal
517
wordsDialogue
251
43.1902%