NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-06-24-01KVX2HQFR0FYWTZE4SYYZFQ3B
Run Details
Model:
Z.AI GLM 5.2 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
11.7s
Cost:
$0.001651
Provider:
Z.AI
Evaluation summary
Score
Evaluator
Details Link
81.27%
Matches word count
(details)
51.03%
Dialogue to Total Word Ratio
(details)
66.1512%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 90% 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
"So here's the plan," Molly said. "We go in through the back gate, blend in with the parents, and start loading presents into the van." "Which parent are we supposed to be?" "The cool ones who brought the bouncy castle." "We didn't bring a bouncy castle." "Exactly. Nobody questions the people who were supposed to bring the bouncy castle. They just assume it's running late." "That's ridiculous." "Trust me, Bill. I've done this before. You carrying the duffel bags?" "Two duffel bags, one fake smile, and a bottle of wine for authenticity." "Perfect. Now when the clown starts his juggling act, that's our window. All eyes on the clown." "What if there's no clown?" "Then we improvise. Maybe you start juggling." "I can't juggle." "Even better. You'll be a distraction. A lovable, incompetent dad just trying his best. Meanwhile, I'm clearing out the gift table like it's a buffet." "What about the piñata?" "What about it?" "Kids go nuts for piñatas, Molly. If that thing breaks while we're loading the van, we're looking at thirty screaming children charging our position." "Fine. You've got five minutes before the candy hits the floor. Move fast." "I always move fast." "You once took twenty minutes to open a pickle jar." "That lid was welded shut."
created_at
"2026-06-24T15:04:02.827Z"
6/24/2026, 3:04:02 PM
model
"z-ai/glm-5.2-20260616"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
4412
moderation_latency
(null)
generation_time
11483
tokens_prompt
77
tokens_completion
414
native_tokens_prompt
73
native_tokens_completion
352
native_tokens_completion_images
(null)
native_tokens_reasoning
53
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
0.001651
router
(null)
provider_responses
0
endpoint_id
"442ea97f-ad5e-40d5-b9a6-66e9e0417dce"
id
"20260624230403fb907e68c2814fc2"
is_byok
false
latency
4412
model_permaslug
"z-ai/glm-5.2-20260616"
provider_name
"Z.AI"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.16.0; linux; x64))"
http_referer
(null)
request_id
"req-1782313442-k486oB8TDBVppJ7yoNT4"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1782313442-fpvEIrVsrRwI2aUOMrO1"
upstream_id
"20260624230403fb907e68c2814fc2"
total_cost
0.001651
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Z.AI"
response_cache_source_id
(null)
data_region
"global"
Evaluation details
Result
Evaluator
Details
Meta Data
81.27%
Matches word count
n/a
neededClean
false
words
212
51.03%
Dialogue to Total Word Ratio
Ratio: 99.06%, Deviation: 9.06%
neededClean
false
wordsTotal
212
wordsDialogue
210
66.1512%