NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYD7RV87DQSDGV395DYVZWVR
Run Details
Model:
MoonshotAI: Kimi K3 (Reasoning, Low)
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
43.4s
Cost:
$0.022484
Provider:
Moonshot AI
Evaluation summary
Score
Evaluator
Details Link
9.61%
Matches word count
(details)
99.99%
Dialogue to Total Word Ratio
(details)
54.7992%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 10% 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 unrolled the blueprint across Bill’s knees, which was difficult because the blueprint was a crayon drawing of a bouncy castle and Bill’s knees were eating crisps. Outside the community center, children shrieked with the sort of joy that made jewel thieves feel underpaid. Inside waited forty presents, one magician, and a piñata shaped like fiscal responsibility. “We go at two, when the clown naps,” Molly whispered. Bill adjusted his disguise, a paper crown reading BIRTHDAY DAD. “And we leave the juice boxes?” “Obviously. We’re monsters, not barbarians.” “Fine.” The plan was elegant: enter through the ball pit, neutralize the bubble machine with gum, compliment the magician until he wept, then slide the gifts down the laundry chute Bill had installed at dawn while pretending to be a raccoon inspector. Molly had packed sacks labeled NOT STOLEN and a getaway wagon disguised as a smaller, guiltier wagon. Bill had packed sandwiches, moral flexibility, and one kazoo for signaling or morale. A toddler wobbled past carrying a gift shaped suspiciously like the diamond tiara from last Easter. Molly narrowed her eyes. The mark had been upgraded from generous to psychotically festive. Bill cracked his knuckles, then apologized to them. Somewhere inside, candles were lit, a choir of sugar went feral, and forty tiny ribbons trembled like hostages who had heard rumors about Molly.
created_at
"2026-07-25T18:14:33.486Z"
7/25/2026, 6:14:33 PM
model
"moonshotai/kimi-k3-20260715"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2110
moderation_latency
(null)
generation_time
43315
tokens_prompt
75
tokens_completion
1491
native_tokens_prompt
147
native_tokens_completion
1496
native_tokens_completion_images
(null)
native_tokens_reasoning
1173
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
0.0224841
router
(null)
provider_responses
0
endpoint_id
"650c9b12-7b6d-444d-8d92-df8daef84d7a"
id
"chatcmpl-6a64fd0a2cae07460ae74526"
is_byok
false
latency
2110
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-1785003273-B9cHIRTVIuPwOxSaTMdV"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1785003273-gwjXwiwwWEVbUiJjCI98"
upstream_id
"chatcmpl-6a64fd0a2cae07460ae74526"
total_cost
0.0224841
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
9.61%
Matches word count
n/a
neededClean
false
words
222
99.99%
Dialogue to Total Word Ratio
Ratio: 9.01%, Deviation: 0.99%
neededClean
false
wordsTotal
222
wordsDialogue
20
54.7992%