NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-06-24-01KVX2Q9VBG5RWYN05ACBHCXCF
Run Details
Model:
Z.AI GLM 5.2 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
1m 17s
Cost:
$0.021389
Provider:
Z.AI
Evaluation summary
Score
Evaluator
Details Link
93.65%
Matches word count
(details)
100.00%
Dialogue to Total Word Ratio
(details)
96.8248%
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
Molly leaned over the kitchen blueprints. "Bill, the target is a six-year-old's birthday party. Blue house on Elm." "Elm Street? What's the security setup?" "A bouncy castle in the backyard and a very loud hired clown." "Armed?" "Helium tank and balloon animals. It's an absolute fortress. What's the haul looking like?" "Thirty wrapped presents. Rumor has it, there's a life-sized Lego castle and a brand new Nintendo Switch." "High stakes. What's our entry point?" Bill whispered, adjusting his dark sweater. "The side gate. It's latched with a standard zip tie." "Child's play. What about timing?" "Exactly fourteen-hundred hours. Right when they cut the cake. The sugar rush creates a tactical blind spot." "And the extraction plan?" "We go in disguised as caterers. We pack the loot into empty pizza boxes and walk right out." "Genius. They'll never suspect the pizza guys. What are the contingency protocols?" "Keep your head on a swivel. If the clown makes direct eye contact, abort immediately." "Why? Is he former military or something?" "No, he's just really good at making aggressive balloon swords." "Understood. Let's move out right away." Molly nodded, packing her tactical wire cutters.
created_at
"2026-06-24T15:07:05.463Z"
6/24/2026, 3:07:05 PM
model
"z-ai/glm-5.2-20260616"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
3611
moderation_latency
(null)
generation_time
76763
tokens_prompt
77
tokens_completion
4201
native_tokens_prompt
73
native_tokens_completion
4838
native_tokens_completion_images
(null)
native_tokens_reasoning
4562
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
0.0213894
router
(null)
provider_responses
0
endpoint_id
"442ea97f-ad5e-40d5-b9a6-66e9e0417dce"
id
"2026062423070567670027f45c47e9"
is_byok
false
latency
3611
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-1782313625-9R4lh1Pe0giyx2kzPbXY"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1782313625-xMVVLhXjTiiEe2JJ7Lu3"
upstream_id
"2026062423070567670027f45c47e9"
total_cost
0.0213894
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
93.65%
Matches word count
n/a
neededClean
false
words
191
100.00%
Dialogue to Total Word Ratio
Ratio: 90.26%, Deviation: 0.26%
neededClean
false
wordsTotal
195
wordsDialogue
176
96.8248%