NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYD7041N0CWJ8DFD9TY1MB9K
Run Details
Model:
MoonshotAI: Kimi K3 (Reasoning, Low)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
32.7s
Cost:
$0.012674
Provider:
Moonshot AI
Evaluation summary
Score
Evaluator
Details Link
20.19%
Matches word count
(details)
0.01%
Dialogue to Total Word Ratio
(details)
10.0993%
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
Molly spread the blueprint across the hood of the borrowed van and anchored it with a juice box. “This,” she said, “is the target.” Bill squinted at the crayon drawing. “That’s a bouncy castle.” “It’s a defensive perimeter.” “It’s got a slide.” “A tactical slide.” Molly tapped the paper with a cupcake she had brought for morale. “The Henderson birthday party begins at two. By two fifteen, every present will be stacked near the gift table while the children are hypnotized by a magician named Gerald the Gentle.” Bill scratched his mustache. “Is Gerald a threat?” “Gerald cried during our last job because someone popped a balloon.” “Good man.” They had robbed banks, museums, and one unfortunate wedding where the ring bearer turned out to be an off-duty detective. But presents were different. Presents were soft targets wrapped in optimism. Molly had spent six weeks training herself to crawl through a ball pit without squeaking. Bill had learned to juggle as a distraction, though mostly he had learned to apologize while juggling. “Entry point?” Bill asked. “Clown car. We arrive disguised as entertainers.” “I’m not wearing big shoes again.” “You’ll wear the shoes and you’ll like the shoes.” “I have a condition.” “You have a dramatic personality.” The party yard beyond the hedge hummed with tiny shoes and unlawful amounts of frosting. A banner read HAPPY 6TH BIRTHDAY POPPY. Balloons bobbed like unsecured thoughts. Somewhere inside, thirty gifts sat wrapped in dinosaurs, fairies, and the kind of silver paper that made Bill emotionally reckless. Molly checked her watch. “We go in, we charm, we load, we leave. No hugging.” “What if a child hugs me first?” “Break the embrace gently.” “What if it’s a very determined child?” “Then suffer.” Bill adjusted his foam nose. “I still think stealing from kids is bad karma.” “We are not stealing joy,” Molly said. “We are reallocating merchandise.” “That sentence has a little mustache of evil on it.” “Thank you.” A clown car rolled up to the curb exactly as planned, except it was occupied by three actual clowns arguing about parking. Molly waved them onward with the serene confidence of a woman holding a balloon sword. The clowns fled. Bill took the wheel, knees around his ears, expression grim. “Remember the signal,” Molly said. “Sneeze twice, drop the cake, shout ‘the raccoon is loose.’” “No. That was the zoo job.” “There was a raccoon.” “There was no raccoon.” “There was in spirit.” They rolled through the gate. Children screamed with delight. Parents took photographs. Gerald the Gentle saw Molly and immediately dropped three doves, none of which had been part of his act. Molly smiled through gritted teeth. “Act natural.” Bill honked the horn by accident, then by commitment. “I am natural.” “You look like a haunted sausage.” “Good,” he said, reaching for the first shiny box. “Then nobody will suspect the sausage.”
created_at
"2026-07-25T18:01:03.296Z"
7/25/2026, 6:01:03 PM
model
"moonshotai/kimi-k3-20260715"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2258
moderation_latency
(null)
generation_time
32625
tokens_prompt
75
tokens_completion
995
native_tokens_prompt
147
native_tokens_completion
842
native_tokens_completion_images
(null)
native_tokens_reasoning
150
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
0.0126741
router
(null)
provider_responses
0
endpoint_id
"650c9b12-7b6d-444d-8d92-df8daef84d7a"
id
"chatcmpl-6a64f9cc24c0ee030681224d"
is_byok
false
latency
2258
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-1785002463-knP6Rymm6vKyaZVZ5dRn"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1785002463-j3SWeiRub1ZslsM2LF0r"
upstream_id
"chatcmpl-6a64f9cc24c0ee030681224d"
total_cost
0.0126741
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
20.19%
Matches word count
n/a
neededClean
false
words
480
0.01%
Dialogue to Total Word Ratio
Ratio: 43.66%, Deviation: 13.66%
neededClean
false
wordsTotal
481
wordsDialogue
210
10.0993%