NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-10-27-01M5YHFWVM5XZMY669NCCKHDMT
Run Details
Model:
Kimi K3.1 Pro (Reasoning, Max)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
3m 13s
Cost:
$0.070572
Provider:
Moonshot AI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
99.99%
Dialogue to Total Word Ratio
(details)
99.9961%
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 blueprint was a paper placemat from Captain Cod’s Fish Bar, and Molly had drawn the bouncy castle too large. “It dominates the whole east wing,” Bill whispered. “It dominates my soul. Focus.” They sat in a van outside 14 Primrose Villas, where seven-year-old Oliver Binks was celebrating his birthday with a jungle theme, a magician, and enough wrapping paper to camouflage a modest tank. Molly lowered her binoculars. Through the kitchen window she could see presents stacked beside a cake shaped like a lion. The lion appeared to be screaming. Bill unfolded a list. Their disguises hung behind him: two catering jackets, one clipboard, and a false moustache large enough to qualify for its own invitation. “We go in as entertainers, lift the presents, and leave during musical statues.” “Why statues?” “Nobody moves. Perfect getaway conditions.” Molly looked at him until he crossed out statues. She revised the plan. “You enter carrying a balloon giraffe, which is thirty-nine uninflated balloons and one bicycle pump. I follow with the cake box, empty except for sacks, tape, and emergency biscuits. At three o’clock, while the magician attempts to locate a volunteer’s dignity, we transfer the presents into the box and wheel it out through the side gate.” “What if the magician sees us?” “Offer him a biscuit.” “What if he wants two?” “Then we negotiate.” A shriek rose from the garden. Bill ducked, but it was only the bouncy castle accepting another child. On the lawn, a woman in a giraffe costume distributed party bags with the weary precision of a prison guard. Molly marked her position, then circled the paddling pool. “No crossing the lawn,” she said. “The grandmother’s watching.” Bill followed her finger. An old woman sat beside the sandbox, motionless except for her eyes, which tracked every adult with bright, financial suspicion. “She looks like security.” “She looks like she invented security.” At two fifty-eight, they put on their jackets. Bill’s moustache slid sideways immediately. Molly straightened it, checked the side gate, and found it locked. Beyond the gate, Oliver’s father began testing a karaoke machine. The first note scattered birds from three gardens. Molly smiled for the first time that afternoon. “Change of plan,” she said. “We go over the fence during ‘Happy Birthday.’ Nobody’ll hear the drill.” Bill stared at her. “We brought a drill?” She patted the cake box. Something inside it whirred, knocked twice, and started to smell of burning biscuits. From the house came a burst of applause. The front door opened. The magician backed out, bowing, while a rabbit climbed determinedly out of his trousers. Molly seized Bill’s sleeve. “Wait. New entrance.” “The rabbit?” “The trousers.” For one hopeful second, neither spoke. Then the bouncy castle exhaled, the lion cake slid onto the patio, and every child in the garden turned toward the van. Bill slowly raised the balloon giraffe. Somewhere behind them, the van’s radio began playing opera at a volume suitable for invasions and natural disasters. Molly sighed. “Start pumping.”
created_at
"2026-10-27T11:21:24.096Z"
10/27/2026, 11:21:24 AM
model
"moonshotai/kimi-k3.1-pro-20261023"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1892
moderation_latency
(null)
generation_time
193370
tokens_prompt
75
tokens_completion
7308
native_tokens_prompt
146
native_tokens_completion
7028
native_tokens_completion_images
(null)
native_tokens_reasoning
6332
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
0.070572
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.22.1; linux; x64))"
http_referer
(null)
request_id
"req-1793100084-fDVz7iPKOuvjJgRel1fs"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1793100084-qyDNA9VIItcz1kvpG0pu"
upstream_id
"chatcmpl-6ae08934216bd289df2176aa"
provider_responses
0
endpoint_id
"a3fafaac-f097-4ef1-86b3-842854dad5c6"
id
"chatcmpl-6ae08934216bd289df2176aa"
is_byok
false
latency
1892
model_permaslug
"moonshotai/kimi-k3.1-pro-20261023"
provider_name
"Moonshot AI"
status
200
total_cost
0.070572
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
99.99%
Dialogue to Total Word Ratio
Ratio: 29.96%, Deviation: 0.04%
neededClean
false
wordsTotal
504
wordsDialogue
151
99.9961%