NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-24-01KQ06XB6ASN1NCWFK4WS885XH
Run Details
Model:
MoonshotAI: Kimi K2.6
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
3m 25s
Cost:
$0.052438
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
100.00%
Dialogue to Total Word Ratio
(details)
100.0000%
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 adjusted night vision goggles over a stolen pastel sundress while Bill strapped a silent vacuum pack to his back. They crouched behind a rhododendron bush overlooking a suburban backyard, where a moonbounce quivered with sugar-fueled chaos and a piƱata hung like a traitor. The gift table sagged beneath Frozen wrapping paper and smelled vanilla icing. A chocolate fountain gurgled nearby, threatening to draw attention from sticky-fingered toddlers. "Grappling hooks?" Bill asked. Molly shook her head, memorizing the layout. "Juice boxes." "Clowns guarding?" "Big red shoes." "Witness policy?" "No one over four feet." "Go time." "Move out." Bill nodded grimly, checking his utility belt for noise-canceling earmuffs. A magician pulled a rabbit from a hat on the deck, providing perfect acoustic cover. The target glittered: thirty-seven unguarded rectangles of plastic wealth guarded only by a deeply sleeping golden retriever wearing a party hat. They synchronized watches as the birthday boy prepared to blow out candles. The distraction would last twelve seconds. Bill tossed confetti grenades into the petunias. The dog twitched but did not wake. Together they crawled past abandoned juice boxes toward untouched Nintendo cartridges and unassembled LEGO empires, ready to redistribute joy in the most felony way possible.
created_at
"2026-04-24T16:59:31.924Z"
4/24/2026, 4:59:31 PM
model
"moonshotai/kimi-k2.6-20260420"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
274
moderation_latency
(null)
generation_time
205248
tokens_prompt
77
tokens_completion
13917
native_tokens_prompt
69
native_tokens_completion
14973
native_tokens_completion_images
(null)
native_tokens_reasoning
13566
native_tokens_cached
32
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"
usage
0.05243805
router
(null)
provider_responses
0
endpoint_id
"a2828210-ff0a-4882-9a4f-cce2aca73946"
id
"chatcmpl-RxVTS2kXCJmK1YwhomnAudYC"
is_byok
false
latency
79
model_permaslug
"moonshotai/kimi-k2.6-20260420"
provider_name
"DeepInfra"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.1; linux; x64))"
http_referer
(null)
request_id
"req-1777049971-V4czf9DuH8UubztpsoLT"
session_id
(null)
api_type
"completions"
id
"gen-1777049971-M1Nbo3hY5nYZTKOLdfDO"
upstream_id
"chatcmpl-RxVTS2kXCJmK1YwhomnAudYC"
total_cost
0.05243805
cache_discount
0.0000192
upstream_inference_cost
0
provider_name
"DeepInfra"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
200
100.00%
Dialogue to Total Word Ratio
Ratio: 9.80%, Deviation: 0.20%
neededClean
false
wordsTotal
204
wordsDialogue
20
100.0000%