NC Bench
Overview
About
Tests
Models
Model
MoonshotAI: Kimi K2.5
Test
Dialogue tags
Scenario
Write 500 words with 70% dialogue
Run ID
2026-02-08-01KGYX7P0YRNF7AQF4CWFZA01A
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 70% 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
Bill spread the blueprint across the dashboard, smoothing crayon wrinkles. "Perimeter's hot. Bounce house blocking east, and the target's mother has hawk eyes." Molly lowered her coffee. "You sketched a heist layout from a Facebook invite?" "Drone surveillance. Gift table's here by the pony. We enter as catering staff." "Hairnets?" "Hairnets are the new ski masks. Nobody suspects the juice box refiller." "They're five years old, Bill." "Age is a number, but an Xbox is four hundred dollars. I've timed the games. Musical chairs ends at 1430. When the music stops, thirty-six sugar-crazed toddlers rush the cake. That's our window." "What about the clown?" "Deniable asset. He does balloons while we hit the garage. I've got a foil-lined duffel for the RFID tags." "You mean price stickers?" "Same technology. We snag the loot, exfil through the ball pit." "Not the ball pit. Last time I found a tooth." "Tic Tac." "It had roots!" "Focus. Reptile show distracts the birthday boy. Twelve minutes until pizza peaks parental supervision." "My grandmother stole diamonds from oligarchs." "She didn't have student loans. Remember the code: if I say 'the unicorn is horny,' abort." "What?" "The cake. If the horn falls, the mother rages. We scatter." "You're ridiculous." "I've practiced my 'fun uncle' smile." "You look like you're having a stroke." "Perfect. Children trust strokes. Check your earpiece." "It's a seashell." "Covert communication. Move out. And if the magician asks you to assist, refuse. I don't trust silk scarves." "Hairnets on?" "Hairnets on. Let's liberate some Nerf guns." "Operation: Goody Bag is a go."
created_at
"2026-02-08T15:18:07.519255+00:00"
Sunday, February 8th, 2026 at 3:18:07 PM GMT+00:00
model
"moonshotai/kimi-k2.5-0127"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
546
moderation_latency
(null)
generation_time
67384
tokens_prompt
61
tokens_completion
2994
native_tokens_prompt
87
native_tokens_completion
2961
native_tokens_completion_images
(null)
native_tokens_reasoning
2882
native_tokens_cached
21
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
usage
0.011155875
router
(null)
provider_responses
0
id
"96e800e021ce45f089c184f9818b37a6"
status
200
is_byok
false
latency
546
endpoint_id
"c884a526-d529-4263-bd90-a58ea25e99d1"
provider_name
"Venice"
model_permaslug
"moonshotai/kimi-k2.5-0127"
api_type
"completions"
id
"gen-1770563819-jQK80MGJaaQl5Ky1g3N1"
upstream_id
"96e800e021ce45f089c184f9818b37a6"
total_cost
0.011155875
cache_discount
0.000013125
upstream_inference_cost
0
provider_name
"Venice"
Evaluation details
Result
Evaluator
Details
Meta Data
0.0000%
Matches word count
n/a
neededClean
false
words
262
0.0000%
Dialogue to Total Word Ratio
Ratio: 94.66%, Deviation: 24.66%
neededClean
false
wordsTotal
262
wordsDialogue
248
0.0000%