NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-10-27-01M5YF0PJGW5SRXBM5KJX2ZMCQ
Run Details
Model:
Kimi K3.1 Pro (Reasoning, Max)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
4m 46s
Cost:
$0.093362
Provider:
Moonshot AI
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 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 birthday invitation across Bill’s kitchen table, pinning one corner with a saltshaker and the other with his unemployed goldfish bowl. “Forty guests, one bouncy castle, and a mountain of presents beside the cake. We go in as magicians, take every present, and leave before the puppet show.” Bill adjusted his paper crown. “Why magicians?” “Because nobody questions a woman carrying a suspiciously large cabinet.” “They will when it’s wheezing.” “That’s your motivation to breathe quietly.” The invitation showed a cartoon badger wearing sunglasses. Beneath it, in green crayon, someone had written: NO CLOWNS. Bill had circled this three times. Molly drew arrows between the back gate, the gift table, and a hedge shaped like a giraffe. Her plan depended on precision, timing, and Bill not speaking to anyone. She had built eleven minutes into the schedule for him to become distracted by cake. Bill studied the diagram with the solemnity of a general invading a very small nation. His disguise lay on the counter: a sequinned jacket, two doves borrowed from a man who believed they were pigeons, and a false moustache previously used in a nativity play. “What’s my stage name?” he asked. “The Great Bill.” “Can’t I be something mysterious?” “Fine. The Great William.” “What about the children?” “What about them?” “They might see us.” “They’ll see two magnificent magicians borrowing their presents for educational purposes.” “Which education?” “Disappointment studies.” “And if one cries?” “Give him a balloon dog.” “I can only make snakes.” “Then make him a long dog and commit completely to the illusion, whatever happens.” Outside, rain ticked against the windows. A car backfired, and one dove applauded. Molly checked her list: rope, sacks, balloons, emergency biscuits. She crossed out rope and wrote MORE BALLOONS. The party started at two. At three, an entertainer called Captain Jellybean would arrive. Molly intended to be gone by then, unless Captain Jellybean had a security background, in which case Bill was instructed to create a diversion by swallowing something inappropriate. “Define inappropriate,” Bill said. “A coin.” “What if he’s impressed?” “Then swallow a larger coin.” Bill nodded, made a note, and ate an emergency biscuit. At half past one, they loaded the cabinet into Molly’s van. It rolled backward, bumped the garden wall, and revealed six feet of empty space where Bill should have been standing. “I’m testing the escape route,” he called from inside. Molly opened the door. Bill sat among the sacks, wearing the moustache on his forehead. They arrived at two. Music thumped beyond the fence. Children shrieked with the pure, piercing joy usually reserved for emergencies. Molly parked beside a sign reading PARTY ENTRANCE and smiled. Then she read the smaller sign below it. TODAY’S PARTY: OFFICER PICKLES’S RETIREMENT. ALL PRESENTS DONATED TO THE POLICE BENEVOLENT FUND. Bill leaned toward the window. “Is that bad?” Molly watched forty uniformed guests converge around the bouncy castle. “No,” she said, handing him the sequinned jacket. “It means you’re definitely doing the coin trick.”
created_at
"2026-10-27T10:38:08.985Z"
10/27/2026, 10:38:08 AM
model
"moonshotai/kimi-k3.1-pro-20261023"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2097
moderation_latency
(null)
generation_time
285464
tokens_prompt
75
tokens_completion
9312
native_tokens_prompt
146
native_tokens_completion
9307
native_tokens_completion_images
(null)
native_tokens_reasoning
8576
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
0.093362
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.22.1; linux; x64))"
http_referer
(null)
request_id
"req-1793097488-FETngqsbCjYdz2mGjUlV"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1793097488-54G18XLbccWxOx8feC17"
upstream_id
"chatcmpl-6ae07f1227d7263b7959f911"
provider_responses
0
endpoint_id
"a3fafaac-f097-4ef1-86b3-842854dad5c6"
id
"chatcmpl-6ae07f1227d7263b7959f911"
is_byok
false
latency
2097
model_permaslug
"moonshotai/kimi-k3.1-pro-20261023"
provider_name
"Moonshot AI"
status
200
total_cost
0.093362
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
100.00%
Dialogue to Total Word Ratio
Ratio: 30.00%, Deviation: 0.00%
neededClean
false
wordsTotal
500
wordsDialogue
150
100.0000%