NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-15-01KKS2R6YD1GVE57EG6ZAN505Y
Run Details
Model:
Z.AI GLM 5 Turbo
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
42.1s
Cost:
$0.019387
Provider:
Z.AI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
100.00%
Dialogue to Total Word Ratio
(details)
100.0000%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 10% 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 unrolled the stolen architectural blueprints across the greasy diner table, her eyes darting toward the schematic of the suburban fortress. Bill nursed his black coffee, staring blankly at the heavily guarded perimeter. The primary obstacle was not the motion sensors or the locked doors, but rather the sheer volume of chaotic toddlers running laps around the property. Balloons shaped like giraffes provided minimal concealment, and the roving security guard was actually a guy in a giant purple dinosaur costume who kept aggressively handing out slices of pizza. Stealing every single birthday present from the backyard required absolute stealth, flawless timing, and a complete lack of shame. Molly tapped the diagram with a greasy French fry, tracing the path from the hedge to the gift table. She calculated the wind speed, the trajectory of flying pizza slices, and the exact seconds before the inevitable sugar crash would incapacitate the juvenile defenders. They needed to strike when the pinata burst, using the resulting shower of Tootsie Rolls as a diversionary tactic to slip past the hyperactive miniature humans and secure loot. "We go in through the bouncy castle. That is a terrible idea. Do you have a better one? Then no."
created_at
"2026-03-15T15:45:02.934Z"
3/15/2026, 3:45:02 PM
model
"z-ai/glm-5-turbo-20260315"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1610
moderation_latency
(null)
generation_time
41900
tokens_prompt
77
tokens_completion
5292
native_tokens_prompt
66
native_tokens_completion
6044
native_tokens_completion_images
(null)
native_tokens_reasoning
5797
native_tokens_cached
22
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
0.019387264
router
(null)
provider_responses
0
endpoint_id
"9ce3b132-3a17-4fea-8092-0e8e519ed32b"
id
"20260315234503b187a8bbbe7e42ca"
is_byok
false
latency
1610
model_permaslug
"z-ai/glm-5-turbo-20260315"
provider_name
"Z.AI"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.0; linux; x64))"
http_referer
(null)
api_type
"completions"
id
"gen-1773589502-6WRMwMOhtgTmhpxggMOt"
upstream_id
"20260315234503b187a8bbbe7e42ca"
total_cost
0.019387264
cache_discount
0.00002112
upstream_inference_cost
0
provider_name
"Z.AI"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
200
100.00%
Dialogue to Total Word Ratio
Ratio: 10.00%, Deviation: 0.00%
neededClean
false
wordsTotal
200
wordsDialogue
20
100.0000%