NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-24-01KQ09S7H9NJ6ETDC16QW2CYCQ
Run Details
Model:
Z.AI GLM 5.1
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
4m 54s
Cost:
$0.061593
Provider:
Phala
Evaluation summary
Score
Evaluator
Details Link
93.65%
Matches word count
(details)
91.08%
Dialogue to Total Word Ratio
(details)
92.3635%
Overall
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 the crayon-drawn map on the sticky laminated diner table in the deserted corner booth, her eyes narrowing at the layout. The purple jelly stain represented the bouncy castle, and the french fry grease smear was the snack table. Bill squinted at the blueprints, scratching his chin and leaving a trail of powdered sugar from his cruller onto the critical infrastructure. "So the primary entry point is the side gate," Molly whispered, glancing around for eavesdroppers. "We go in during the piñata. It’s the optimal window. Total sensory overload." "I don't know, Moll. The piñata is high risk. What if the kid gets a lucky swing? My cousin lost a molar that way." She rolled her eyes so hard it looked physically painful and pointed the blue crayon at a square labeled Loot Drop. She tapped it twice for emphasis, leaving a waxy smudge on the cheap paper napkin. "Focus. While the toddlers are beating a paper donkey, we hit the present table. You take the shiny bags, I take the boxes." "What if they used that impossible Scotch tape? I hate it. My thumbs get red." Molly sighed deeply, rubbing her temples with both hands as if trying to keep her brain from exploding out of her ears while the fluorescent lights buzzed overhead. She took a slow, calming breath. "We don't unwrap them on site, Bill. We take the haul back to the safe house. Security is light but crafty. We have a bouncy castle monitor and a mom on photo duty." "What about the clown? We didn't factor in the clown." Bill shuddered visibly at the memory of a previous circus job. Molly consulted her napkin notes, adjusting her reading glasses and ignoring his trembling shoulders. She traced a line around the perimeter of the yard. "The clown is on a candy route. He’ll be making balloon animals by the sandbox. If you avoid eye contact, you're invisible." "I can do invisible. What's my cover if I get cornered by a parent?" She slid a crumpled paper bag across the table, the rustling sound loud in the empty diner. Her expression was completely serious. "Put on the apron. You're the caterer. If anyone asks, you're refilling the juice boxes. Just look stressed and complain." "I can do stressed. This is a big job, Molly. These are ruthless six-year-olds. They absolutely bite." Molly snapped her crayon in half, leaning across the table with a manic gleam in her eye, glaring at him. She grabbed Bill by the collar of his jacket, pulling him close enough to smell the donut. "Stick to the plan. We infiltrate, execute, exfiltrate. Nobody gets bitten, and we walk out with every Lego set and gift card. Are you in?" "I'm in. But if the dog chases me, I'm dropping the presents and climbing a tree." She nodded firmly, tucking the crumpled map into her leather jacket and leaving a crumpled dollar tip on the table. It was time to bring home the real birthday haul. "Deal. Eat, we move at noon."
created_at
"2026-04-24T17:49:42.836Z"
4/24/2026, 5:49:42 PM
model
"z-ai/glm-5.1-20260406"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1914
moderation_latency
(null)
generation_time
293127
tokens_prompt
77
tokens_completion
12377
native_tokens_prompt
66
native_tokens_completion
14646
native_tokens_completion_images
(null)
native_tokens_reasoning
11620
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"
usage
0.06159306
router
(null)
provider_responses
0
endpoint_id
"821c357a-76d4-4ec0-a0fa-0740d54aefec"
id
"2fdccde3276940398ca3ca2a38e38dcd"
is_byok
false
latency
1914
model_permaslug
"z-ai/glm-5.1-20260406"
provider_name
"Phala"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.1; linux; x64))"
http_referer
(null)
request_id
"req-1777052982-Jjc1JIxvdHrc6yjZSYSA"
session_id
(null)
api_type
"completions"
id
"gen-1777052982-6NUxGe8Sj2ur1Hq1NW8w"
upstream_id
"2fdccde3276940398ca3ca2a38e38dcd"
total_cost
0.06159306
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Phala"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
93.65%
Matches word count
n/a
neededClean
false
words
509
91.08%
Dialogue to Total Word Ratio
Ratio: 48.63%, Deviation: 1.37%
neededClean
false
wordsTotal
512
wordsDialogue
249
92.3635%