NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-06-24-01KVX3E47P9W4W78K1GRDA7ERC
Run Details
Model:
Z.AI GLM 5.2 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
1m 38s
Cost:
$0.030581
Provider:
Z.AI
Evaluation summary
Score
Evaluator
Details Link
99.74%
Matches word count
(details)
83.38%
Dialogue to Total Word Ratio
(details)
91.5623%
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 crayon-drawn map of the Henderson backyard across the sticky kitchen table. Bill leaned over it, a half-eaten jelly donut in one hand, leaving powdered sugar across the escape route. The air in their safe house smelled of stale coffee and desperation. A flickering light cast long shadows over their faces as they prepared for the most ridiculous job of their careers. "The target is a six-year-old," Molly said, tapping the map with a pencil. "He is having a dinosaur-themed birthday party at fourteen hundred hours." Bill snorted, wiping donut glaze on his shirt. He asked what the security could be for a toddler. A baby gate? He tossed a crumb onto the paper, mocking the operation. Molly gave him a cold stare that shut him up. "Do not underestimate the Hendersons," Molly warned, her eyes narrowing. "They have a bouncy castle. And they hired a clown." Bill shuddered visibly. He had a bad history with clowns, mostly involving a mishap at a county fair and a fireworks display that left him with a lifelong phobia of red noses. He swallowed hard and inquired about the entry point. Molly traced a line along the side garden. She explained the parents would be distracted by a magician at two-fifteen. Bill would create a diversion near the piƱata while she slipped into the garage to secure the loot. When he asked what kind of diversion she had in mind, her response made his blood run cold. "You are going to pop the bouncy castle." Bill dropped his donut, watching it roll under the radiator. He protested they were thieves, not monsters. He argued the trauma of an inflating dinosaur collapsing onto six-year-olds would bankrupt their therapy funds. Molly ignored his objections. She reminded him of the battery-powered jeep among the presents. Fencing that item would pay off the loan shark threatening to remove their fingers. "Pop the castle? We are thieves, not monsters. Think of the therapy bills." "It is a calculated risk," Molly replied smoothly. "We slice the base. The deflation draws the adults. We are in and out. There is a mountain of toys. We fence the jeep and pay the loan shark." Bill picked the fallen donut up, blowing dust off the icing. He considered the moral weight of destroying a child's inflatable joy against the threat of having his kneecaps shattered with a baseball bat. The scales tipped in favor of his kneecaps. "Fine," Bill muttered. "I slice the castle. But what if the clown sees me?" "The clown is a non-issue. I spiked his water with a laxative this morning. He is out of commission." Bill grinned, a genuine smile breaking through his stubble. He told her she was a terrifying woman. "You are a terrifying woman, Molly." "I am a thorough woman," she corrected. "Grab the bags. We leave in ten minutes. Try not to eat the cake on the way out." "No promises," Bill said. "If I am deflating a bouncy T-Rex, I am taking a cupcake."
created_at
"2026-06-24T15:19:33.38Z"
6/24/2026, 3:19:33 PM
model
"z-ai/glm-5.2-20260616"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2833
moderation_latency
(null)
generation_time
97317
tokens_prompt
77
tokens_completion
6507
native_tokens_prompt
73
native_tokens_completion
6927
native_tokens_completion_images
(null)
native_tokens_reasoning
6286
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
0.030581
router
(null)
provider_responses
0
endpoint_id
"442ea97f-ad5e-40d5-b9a6-66e9e0417dce"
id
"20260624231933380c6c6d87b74125"
is_byok
false
latency
2833
model_permaslug
"z-ai/glm-5.2-20260616"
provider_name
"Z.AI"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.16.0; linux; x64))"
http_referer
(null)
request_id
"req-1782314373-p8glcNeK1sfst0llmEtw"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1782314373-lsCqs1LoiJoTYyixKHmK"
upstream_id
"20260624231933380c6c6d87b74125"
total_cost
0.030581
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Z.AI"
response_cache_source_id
(null)
data_region
"global"
Evaluation details
Result
Evaluator
Details
Meta Data
99.74%
Matches word count
n/a
neededClean
false
words
504
83.38%
Dialogue to Total Word Ratio
Ratio: 31.91%, Deviation: 1.91%
neededClean
false
wordsTotal
514
wordsDialogue
164
91.5623%