NC Bench
Overview
About
Tests
Models
Model
Z.AI GLM 4.5
Test
Dialogue tags
Scenario
Write 200 words with 90% dialogue
Run ID
2026-02-08-01KGYXQPTR136KZ7JVBE9DZAXV
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 90% 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
“Are you sure this is the right address, Bill? It smells suspiciously like joy and frosting.” “Positive, Molly. 321 Sugarplum Lane. Intel says the presents are in the master bedroom, away from the sticky-fingered guests.” “Sticky-fingered guests? Bill, we’re about to burgle a six-year-old’s birthday party. Our moral compass is spinning like a dreidel.” “Focus! The parents are distracted by a man in a giant bear costume. It’s the perfect window of opportunity.” “A window? I was thinking of just walking through the front door. No one suspects a guest at a child’s party.” “They will when you try to leave with an armful of wrapped gifts. We go in through the den window. It’s always unlocked.” “Fine. But if I see a piñata, I’m taking a swing at it. For research purposes.” “Molly, for the love of loot, just stick to the plan. Get in, grab the goods, get out. No detours, no candy.” “You’re no fun. Okay, lead on, Captain Grump. Let’s go ruin some childhoods.”
created_at
"2026-02-08T15:25:50.999546+00:00"
Sunday, February 8th, 2026 at 3:25:50 PM GMT+00:00
model
"z-ai/glm-4.5"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1476
moderation_latency
(null)
generation_time
5336
tokens_prompt
61
tokens_completion
240
native_tokens_prompt
70
native_tokens_completion
244
native_tokens_completion_images
0
native_tokens_reasoning
0
native_tokens_cached
24
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
0.00056704
router
(null)
provider_responses
0
id
"a887d55026d71b6d398df09832c9200e"
status
200
is_byok
false
latency
1476
endpoint_id
"e4ada6ec-12e7-49c9-837c-398dee612403"
provider_name
"Novita"
model_permaslug
"z-ai/glm-4.5"
api_type
"completions"
id
"gen-1770564344-sklfbkPg644irXcXRbyV"
upstream_id
"a887d55026d71b6d398df09832c9200e"
total_cost
0.00056704
cache_discount
0.00001176
upstream_inference_cost
0
provider_name
"Novita"
Evaluation details
Result
Evaluator
Details
Meta Data
0.0848%
Matches word count
n/a
neededClean
false
words
171
36.7879%
Dialogue to Total Word Ratio
Ratio: 100.00%, Deviation: 10.00%
neededClean
false
wordsTotal
171
wordsDialogue
171
18.4364%