NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-02-01M1GTDE37JW7S4CG55TW5W07B
Run Details
Model:
Z.AI GLM 5.3 (Reasoning, Max)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
4m 33s
Cost:
$0.077476
Provider:
Z.AI
Evaluation summary
Score
Evaluator
Details Link
0.49%
Matches word count
(details)
99.98%
Dialogue to Total Word Ratio
(details)
50.2380%
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 crouched behind the hedges of number fourteen, binoculars pressed to her face, a hand-drawn map of the suburban semi tucked under one arm like a sacred text. Beside her, Bill lay flat in the grass, wearing a camouflage jacket over his gardening trousers, holding a clipboard for no reason anyone could explain. "Target acquired," she whispered. "Bouncy castle in the rear garden. Twelve hostiles, ages five to seven." Bill made a note. A shrieking child rocketed past in a paper crown, and both thieves instinctively ducked as though it were tracer fire. "Entertainment?" he asked. "Clown. Balloon animals indicate professional training." He nodded gravely. They peered over the hedge at the party in all its terrifying glory: streamers, a piñata dangling from the washing line like a hanged informant, and a folding table groaning under a battalion of brightly wrapped boxes. "The present table is by the patio doors," Molly continued. "Nine gifts. One of them is castle-shaped, which is a minimum of forty quid at retail." Bill studied the scene, chewing his pen. He looked less like a master criminal and more like a man reconsidering his mortgage. He asked about security. "Mother at the door. Sharp eyes. She clocked the postman at forty metres, and he's been coming here twice a week for years." "Dog?" asked Bill. "Dachshund. Don't laugh. That's a low-profile alarm system with teeth." Bill was already laughing and took an elbow in the ribs for it. They retreated to the van, a beige Transit that had once said PLUMBER on the side and now just said PLUM, and spread the map across the dashboard. Molly tapped the garden with one finger, delivering the plan she'd rehearsed in the bath that morning. They would slip through the side gate at fifteen hundred hours, the moment the clown began his finale, when every eye would be upon him and whatever indignity he planned to inflict upon a balloon. "And the mother?" Bill asked. "She'll have her hands full with cake. Nobody watches presents when there's cake. It's the oldest distraction in the book. I invented it." "You didn't invent cake." "I invented using it." Bill stared at the map, which was actually a flyer for the local chip shop with a garden drawn on it in biro. He chewed his pen. "You're sure about this?" he said at last. "We've done banks. We've done galleries. None of it made me this nervous. Kids notice things, Molly. Kids notice everything." "Then we're in and out in four minutes. Grab, bag, gone." "And the piñata?" Bill asked, and when she demanded to know what about it, he said only that it had been staring at him all afternoon, like it knew. Molly snatched up the binoculars and fixed her gaze on the castle-shaped box, resplendent in silver paper, sitting on the table like a crown jewel of commerce. Somewhere beyond the hedge a child began to wail about a juice box, and she felt the old familiar thrill, the tingle of a job about to go loud. "Fifteen hundred hours," she said. "God help us," said Bill, and made a note of that too.
created_at
"2026-09-02T10:25:11.279Z"
9/2/2026, 10:25:11 AM
model
"z-ai/glm-5.3-20260816"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2787
moderation_latency
(null)
generation_time
273201
tokens_prompt
75
tokens_completion
15066
native_tokens_prompt
73
native_tokens_completion
17585
native_tokens_completion_images
(null)
native_tokens_reasoning
16889
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
0.0774762
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788344711-XxTe7DqpUH6c79iRJFRs"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788344711-ucAXM4mnlyusLaCv6Cy1"
upstream_id
"2026090218251128c7141af9d64073"
provider_responses
0
endpoint_id
"db975786-e7f6-4838-9e56-3e0b77f1bdc2"
id
"2026090218251128c7141af9d64073"
is_byok
false
latency
2787
model_permaslug
"z-ai/glm-5.3-20260816"
provider_name
"Z.AI"
status
200
total_cost
0.0774762
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
0.49%
Matches word count
n/a
neededClean
false
words
527
99.98%
Dialogue to Total Word Ratio
Ratio: 29.94%, Deviation: 0.06%
neededClean
false
wordsTotal
531
wordsDialogue
159
50.2380%