NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-02-01M1H576FKBFGMHV0A3AMW8K8D
Run Details
Model:
Z.AI GLM 5.3 Flash (Reasoning, Low)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
22.0s
Cost:
$0.000182
Provider:
Z.AI
Evaluation summary
Score
Evaluator
Details Link
99.98%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
49.9920%
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
The blueprints of the cardboard castle lay spread across Bill's kitchen table, held down by a coffee mug and a bag of gummy bears that Molly kept picking at. "Explain to me again why we're robbing a seven-year-old's birthday party," Bill said, rubbing his temples. Molly tapped the diagram with a pencil. "Because little Tommy Pemberton is getting a seventy-five pound toy robot, a bicycle, and according to my source at the bakery, a solid chocolate treasure chest. That's two grand in loot, minimum." "It's a birthday party, Molly. There will be parents there. Vigilant, caffeinated parents." "That's why we go as clowns." Bill stared at her. "Absolutely not." "Bill, think about it. We blend in. Balloon animals are the perfect distraction technique. I do the twisty poodle thing, you make a giraffe, and while the children scream with delight, we wheel the present table out the fire exit." "First of all, I don't make balloon animals. Second of all, I've got a face for hiding in shadows, not for clown makeup." "You've got a face that says 'trustworthy neighborhood clown,' which is honestly worse, but we'll work with it." She rolled up the blueprint and smacked him on the arm with it. "The party's Saturday, two to five p.m., Killarney Community Hall. Hiring a professional clown costs three hundred pounds, so nobody will blink when one shows up. Well, they'll blink at your nose." "What about the magician?" Molly paused. "There's a magician?" "Tommy's mum hired Magic Marco. I read the Facebook page. You said to do research, so I did research." "Marco." Molly's eyes narrowed dangerously. "Marco owes me money from a vending machine scheme in 2019. If he's there, this gets complicated." "Or," Bill said slowly, "we could just... not rob children." "Bill." "I'm just saying. We've robbed a casino, two jewelry stores, and a man who turned out to be an off-duty bishop. None of those involved party bags." "The bishop had party bags." "He had a spare mitre in his car, that's not the same." Bill sighed, looking at the blueprint again. "The robot alone. How much did you say?" "Seventy-five pounds retail. Plus the bike, plus whatever's in those pastel gift bags. We split it sixty-forty because I did the recon." "You always do the recon and always say sixty-forty." "And you always accept, because you have no head for logistics." Bill chewed a gummy bear, contemplating his life choices. "Fine. Saturday. But if I'm doing clown duty, I want the red nose with the squeaker. The squeaky one." "That's the professional model. Costs extra." "So does the robot, Molly." She grinned and pulled a small box from her bag, sliding it across the table. A squeak emerged. "God help me," Bill muttered, trying it on. "How do I look?" Molly studied him for a long moment. "Like a man about to steal presents from children at a birthday party." "Great." "Which is exactly what a trustworthy neighborhood clown should look like. We leave at one."
created_at
"2026-09-02T13:34:01.215Z"
9/2/2026, 1:34:01 PM
model
"z-ai/glm-5.3-flash-20260826"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1166
moderation_latency
(null)
generation_time
21961
tokens_prompt
75
tokens_completion
766
native_tokens_prompt
73
native_tokens_completion
705
native_tokens_completion_images
(null)
native_tokens_reasoning
10
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
0.000181725
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788356041-BM5lV8qoUVW1UMorX4E5"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788356041-gPjZzatrxwwV4O92yrXK"
upstream_id
"20260902213401bc80ed21981b4056"
provider_responses
0
endpoint_id
"8e9fe48b-2f91-41c3-a8a7-e4a93a8c4ff0"
id
"20260902213401bc80ed21981b4056"
is_byok
false
latency
1166
model_permaslug
"z-ai/glm-5.3-flash-20260826"
provider_name
"Z.AI"
status
200
total_cost
0.000181725
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Z.AI"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
99.98%
Matches word count
n/a
neededClean
false
words
502
0.00%
Dialogue to Total Word Ratio
Ratio: 77.80%, Deviation: 47.80%
neededClean
false
wordsTotal
509
wordsDialogue
396
49.9920%