NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-02-01M1GQM36KGVJBVJ24ECTHHVVM
Run Details
Model:
Z.AI GLM 5.3 (Reasoning, Max)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
2m 37s
Cost:
$0.046465
Provider:
Z.AI
Evaluation summary
Score
Evaluator
Details Link
95.99%
Matches word count
(details)
2.09%
Dialogue to Total Word Ratio
(details)
49.0399%
Overall
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 kitchen table had been cleared of everything except a thermos, two mugs, and a floor plan of a suburban back garden drawn in purple felt tip. Molly tapped the paddling pool with a wooden spoon. "Okay. Sunday. Fourteen hundred hours. The bouncy castle arrives at one. That's our window." "The bouncy castle is a window?" "The bouncy castle is a distraction, Bill. Keep up. The presents go on this table here. Spider-Man theme. I counted twelve gift bags, three wrapped boxes, and something large behind the shed that could be a scooter." "That's the getaway vehicle sorted, then." "That's a child's scooter. It has streamers." "Streamers are festive." Bill studied the plans with the focus of a man defusing a bomb. "What about the clown?" "What clown?" "Party entertainer. Face painting, balloon animals. Word on the street is he does close-up magic." "Since when is there word on the street about a birthday clown?" "Since I did my own recon. His name is Mr. Twisters. Five-star reviews." "Are you reviewing children's entertainers online?" "I'm thorough. If he spots us, he'll distract us with a card trick while the mums call the police." "Nobody's calling the police. It's a seven-year-old's party." "Then who do they call?" "Each other. Loudly. With WhatsApp voice notes." Bill shuddered. Those could go on for minutes. "Which is why we're in and out in six. Disguises are in the bag. Matching lanyards. Catering staff. Nobody questions a lanyard." "We're not catering." "We're catering staff who've finished catering. We're collecting the tableware, and the tableware happens to be sitting under the presents. Two birds." "What about the piñata? Do we take the piñata?" "It's full of Haribo." "...I'll fetch the bin bags." Molly pinched the bridge of her nose, a gesture stolen from every detective show ever made. "One rule. The cake. If anything happens to the cake—" "You'll leave me at the party?" "I will leave you at the party, and I will tell Mr. Twisters you want a balloon octopus." Bill went pale. "Those take him ages." "Then we're aligned. Now, perimeter. Side gate by the hydrangeas, plus a nosy neighbour at number forty-two. Mrs. Penrose. She watches Gardeners' World with binoculars." "Binoculars for gardening?" "She claims it's for the birds. It is not for the birds." "So we neutralise her." "We wave. We smile. We're catering staff. Catering staff wave." He nodded slowly, then paused. "And the getaway?" "Round the corner, into the van, drive calmly at the speed limit." "We could take the scooter. For morale." Molly considered it. The scooter did have very good streamers. "Fine. The scooter's for morale. But if a single child cries—" "There's always one who loses it over pass the parcel anyway." "That's true. There's always one." "Statistically, it's usually the birthday kid." Molly rolled up the blueprint and tucked the wooden spoon into her jacket like a sword. "Sunday, then. Lanyards on, cake intact, out before Mr. Twisters can ask us to pick a card." Bill saluted. "For the presents." "For the presents."
created_at
"2026-09-02T09:36:23.77Z"
9/2/2026, 9:36:23 AM
model
"z-ai/glm-5.3-20260816"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
3083
moderation_latency
(null)
generation_time
157158
tokens_prompt
75
tokens_completion
9877
native_tokens_prompt
73
native_tokens_completion
10537
native_tokens_completion_images
(null)
native_tokens_reasoning
9801
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
0.046465
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788341783-jrBmL0scdNNMBhsFQEkO"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788341783-jvddZ5ggbqivtv3uWYLr"
upstream_id
"20260902173623ae2dd5b3da56436c"
provider_responses
0
endpoint_id
"db975786-e7f6-4838-9e56-3e0b77f1bdc2"
id
"20260902173623ae2dd5b3da56436c"
is_byok
false
latency
3083
model_permaslug
"z-ai/glm-5.3-20260816"
provider_name
"Z.AI"
status
200
total_cost
0.046465
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
95.99%
Matches word count
n/a
neededClean
false
words
508
2.09%
Dialogue to Total Word Ratio
Ratio: 78.79%, Deviation: 8.79%
neededClean
false
wordsTotal
514
wordsDialogue
405
49.0399%