NC Bench
Overview
About
Tests
Models
Model
Z.AI GLM 4.6
Test
Dialogue tags
Scenario
Write 500 words with 50% dialogue
Run ID
2026-02-08-01KGYZ2B99SCG49XWFHKEHE3JQ
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 50% 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 tapped a greasy finger on the blueprint, which was actually a ketchup-smeared placemat from “Pizza Pete’s.” Across the van’s worn bench seat, Bill nervously picked at a loose thread on the upholstery, the scent of stale pepperoni filling the small space. The placemat showed a crude, red-squiggled map of a suburban backyard. “Entry point is the side gate, left unlocked by the older sibling for a five-dollar bribe,” Molly said, her voice low and serious. “We move along the fence line, using the hedge as cover. Our target, the present table, is here.” She pointed to a large red blob. “Guarded by the primary parental unit, Brenda.” Bill squinted at the schematic. “What about the bouncy castle? Those things are unpredictable. A kid on a sugar high could launch into me at any moment. It’s a rogue projectile.” Molly sighed, a long-suffering sound that seemed to absorb all the van’s ambient light. “The bouncy castle is a non-target asset, Bill. We treat it like a pressure plate: you don’t touch it. My main concern is the clown.” “‘Silly Steve’?” Bill asked, his voice rising slightly. “He’s a wildcard. He might squirt me with a flower or make me a balloon animal, compromising my identity.” “We neutralize Steve with a riddle,” Molly countered, not missing a beat. “He’s a sucker for a good brain-teaser. It will create a ten-minute window of confusion. The real threat is the glitter.” “The glitter?” “It’s a contaminant, Bill. A forensic nightmare. It gets on everything. We’ll need lint rollers for the extraction.” She paused, her eyes narrowing. “And what if a kid gets face paint on my tactical turtleneck? That’s dry-clean only.” Bill stared at her, then back at the placemat. He couldn’t believe his life had come to this. “You know, for a master jewel thief, you’re surprisingly worried about a little craft glitter.” Molly leaned forward, her professional facade cracking just enough to reveal a glint of childish glee. “I saw one of the gift bags through the binoculars. It had a limited-edition, holographic, sparkly unicorn on it. I’m not leaving without it.”
created_at
"2026-02-08T15:50:20.66978+00:00"
Sunday, February 8th, 2026 at 3:50:20 PM GMT+00:00
model
"z-ai/glm-4.6"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
746
moderation_latency
(null)
generation_time
78185
tokens_prompt
61
tokens_completion
2122
native_tokens_prompt
66
native_tokens_completion
2145
native_tokens_completion_images
(null)
native_tokens_reasoning
1727
native_tokens_cached
64
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
0.0037382799999552
router
(null)
provider_responses
0
id
"chatcmpl-Rcel5FXzhpgndweTfAAZ5iE7"
status
200
is_byok
false
latency
59
endpoint_id
"c8607edc-8577-4199-b03d-1213da4743dc"
provider_name
"DeepInfra"
model_permaslug
"z-ai/glm-4.6"
api_type
"completions"
id
"gen-1770565741-qP4k0ngX0yJTx3IhCK3u"
upstream_id
"chatcmpl-Rcel5FXzhpgndweTfAAZ5iE7"
total_cost
0.0037382799999552
cache_discount
0.0000224000000448
upstream_inference_cost
0
provider_name
"DeepInfra"
Evaluation details
Result
Evaluator
Details
Meta Data
0.0000%
Matches word count
n/a
neededClean
false
words
360
0.1108%
Dialogue to Total Word Ratio
Ratio: 61.67%, Deviation: 11.67%
neededClean
false
wordsTotal
360
wordsDialogue
222
0.0554%