NC Bench
Overview
About
Tests
Models
Model
Z.AI GLM 4.6
Test
Dialogue tags
Scenario
Write 500 words with 70% dialogue
Run ID
2026-02-08-01KGZ053XGYEZNWBEBA8TNPF36
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
The vinyl of the diner booth stuck to the back of Molly’s thighs. She slid a crumpled napkin across the table. On it, a crude diagram of a suburban backyard, complete with a gift table circled in red. “Okay, run it back one more time,” she said, her voice low. “In and out, ten minutes, tops.” Bill, using a french fry as a pointer, tapped the drawing. “Ten minutes? But what about the bouncy castle? You can’t rush a bouncy castle experience, Molly. It’s about the journey, not the destination.” “We are not getting in the bouncy castle, Bill. We’re there for the presents. The payload. The brightly wrapped loot.” “Right, right, the payload.” He took a bite of the fry. “But the presents are *for* the kids. Are we the bad guys in this scenario? I feel like we’re the bad guys.” Molly rubbed her temples. “We’re professional thieves, Bill. Our moral compass spins like a roulette wheel. Now, focus. The diversion.” “The piñata!” Bill’s eyes lit up. “I was thinking, what if I dressed up as the piñata? They’d never see it coming. I’d just… stand there. Absorb the anticipation. Then, when the time is right, I burst open, but instead of candy, it’s smoke bombs!” “No. You are not dressing up as a piñata,” Molly said, gripping her coffee mug. “We stick to the plan. I create a distraction by ‘winning’ the three-legged race. While the parents are fawning over my athletic prowess, you sneak the gift table into the van.” “But what if I win the three-legged race? I’ve been practicing.” “You can’t even walk and chew gum at the same time, Bill. Your job is to lift. My job is to be vaguely sporty.” “Okay, okay…” He leaned back, the booth groaning in protest. “But if there’s a clown, I get to steal his nose. That’s non-negotiable.” Molly stared at him for a long moment, then sighed. “Fine. You can steal the stupid nose. Now, pass the ketchup. This meatloaf isn’t going to pretend to be edible by itself.”
created_at
"2026-02-08T16:09:19.396993+00:00"
Sunday, February 8th, 2026 at 4:09:19 PM GMT+00:00
model
"z-ai/glm-4.6"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
349
moderation_latency
(null)
generation_time
77851
tokens_prompt
61
tokens_completion
1942
native_tokens_prompt
66
native_tokens_completion
1957
native_tokens_completion_images
(null)
native_tokens_reasoning
1521
native_tokens_cached
32
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
0.0034223599999776
router
(null)
provider_responses
0
id
"chatcmpl-RoHaflkxZHE3uEcdCifDm7J1"
status
200
is_byok
false
latency
56
endpoint_id
"c8607edc-8577-4199-b03d-1213da4743dc"
provider_name
"DeepInfra"
model_permaslug
"z-ai/glm-4.6"
api_type
"completions"
id
"gen-1770566881-uA17R7Xu8kTtMRrnlIKl"
upstream_id
"chatcmpl-RoHaflkxZHE3uEcdCifDm7J1"
total_cost
0.0034223599999776
cache_discount
0.0000112000000224
upstream_inference_cost
0
provider_name
"DeepInfra"
Evaluation details
Result
Evaluator
Details
Meta Data
0.0000%
Matches word count
n/a
neededClean
false
words
347
59.9517%
Dialogue to Total Word Ratio
Ratio: 73.20%, Deviation: 3.20%
neededClean
false
wordsTotal
347
wordsDialogue
254
29.9758%