NC Bench
Overview
About
Tests
Models
Model
Z.AI GLM 4.7
Test
Dialogue tags
Scenario
Write 200 words with 90% dialogue
Run ID
2026-02-08-01KGZ00R7DVS3452G23B8ZNAB0
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
"Target confirmed," Molly hissed, checking the perimeter. "The bouncy castle is unguarded." Bill adjusted his ski mask, looking miserable. "We’re really doing this? Robbing a six-year-old? The security is a single dad manning a grill." "It's a tactical barbecue distraction," Molly snapped. "Focus. The piñata is the primary vault. We breach the donkey, we secure the payload." "The payload is Tootsie Rolls, Molls. And maybe a plastic spider ring." "That ring is high-yield liquidation! Now, what’s the entry vector?" "The gate is open." "Too obvious. We deploy through the ball pit." Bill groaned. "I lost a shoe in a ball pit in '09. I’m still dealing with the PTSD." "Suck it up. If the clown gets hostile, use the balloon animals as a garrote. I need crowd control while I sweep the gift table." "This is insane. What if a kid cries?" "Then you offer him a distraction. Tell him you saw a unicorn." "A unicorn?" "Just do it! We move when they sing 'Happy Birthday'. The audio masking will be perfect." "Fine. But if I get glitter on this suit, I’m keeping the ransom money."
created_at
"2026-02-08T16:06:52.612999+00:00"
Sunday, February 8th, 2026 at 4:06:52 PM GMT+00:00
model
"z-ai/glm-4.7-20251222"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1167
moderation_latency
(null)
generation_time
74063
tokens_prompt
61
tokens_completion
1386
native_tokens_prompt
66
native_tokens_completion
1391
native_tokens_completion_images
(null)
native_tokens_reasoning
1127
native_tokens_cached
2
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
0.00309882
router
(null)
provider_responses
0
id
"202602090005385145bbc6a5a24ebe"
status
200
is_byok
false
latency
1167
endpoint_id
"be4acbf5-9fcf-4332-a01d-76dfeb6d7b99"
provider_name
"Z.AI"
model_permaslug
"z-ai/glm-4.7-20251222"
api_type
"completions"
id
"gen-1770566738-UvtCiLYEFgbdx7vPc04y"
upstream_id
"202602090005385145bbc6a5a24ebe"
total_cost
0.00309882
cache_discount
9.8e-7
upstream_inference_cost
0
provider_name
"Z.AI"
Evaluation details
Result
Evaluator
Details
Meta Data
86.3803%
Matches word count
n/a
neededClean
false
words
189
99.9446%
Dialogue to Total Word Ratio
Ratio: 91.53%, Deviation: 1.53%
neededClean
false
wordsTotal
189
wordsDialogue
173
93.1625%