NC Bench
Overview
About
Tests
Models
Model
Z.AI GLM 4.7
Test
Dialogue tags
Scenario
Write 200 words with 90% dialogue
Run ID
2026-02-08-01KGYZ3P21QDC1WNWHJAEWXY0B
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 is confirmed," Bill whispered, scanning the yard. "Six-year-old’s birthday party. High sugar intake, low attention spans." "Excellent," Molly replied, checking her watch. "What’s the sit-rep on the primary obstacle?" "The bouncy castle?" "No, the parents. Specifically the one with the camera." "Neutralized. They’re busy arguing about gluten-free options. We have a thirty-second window on the gift table." "Thirty seconds? That’s tight. Are we facing the plastic tablecloth or the paper?" "Paper with ducks. Static electricity is a factor." "Damn it, Bill, I told you to bring the anti-static spray. We’ll be stuck to the wrapping paper." "It wasn't on the list! The list said 'bolt cutters'!" "For the ribbons! You can’t just untie a professional bow in a heist, you have to slice it." "Okay, look, focus. The bicycle is the crown jewel. We grab the Huffy, we ignore the clothes." "What about the cards? There’s cash in those." "Too risky. Opening a card takes manual dexterity. We stick to the boxed goods. Move on my signal." "What’s the signal?" "When the clown drops a balloon." "He drops them constantly!" "Exactly. It’s go time."
created_at
"2026-02-08T15:50:15.752854+00:00"
Sunday, February 8th, 2026 at 3:50:15 PM GMT+00:00
model
"z-ai/glm-4.7-20251222"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
9552
moderation_latency
(null)
generation_time
29610
tokens_prompt
61
tokens_completion
1659
native_tokens_prompt
66
native_tokens_completion
1669
native_tokens_completion_images
(null)
native_tokens_reasoning
1395
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
0.00371042
router
(null)
provider_responses
0
id
"2026020823494548ea561601364f11"
status
200
is_byok
false
latency
9552
endpoint_id
"be4acbf5-9fcf-4332-a01d-76dfeb6d7b99"
provider_name
"Z.AI"
model_permaslug
"z-ai/glm-4.7-20251222"
api_type
"completions"
id
"gen-1770565785-pVa037gZHsN65zUGyMCG"
upstream_id
"2026020823494548ea561601364f11"
total_cost
0.00371042
cache_discount
9.8e-7
upstream_inference_cost
0
provider_name
"Z.AI"
Evaluation details
Result
Evaluator
Details
Meta Data
90.4837%
Matches word count
n/a
neededClean
false
words
190
95.0901%
Dialogue to Total Word Ratio
Ratio: 94.74%, Deviation: 4.74%
neededClean
false
wordsTotal
190
wordsDialogue
180
92.7869%