NC Bench
Overview
About
Tests
Models
Model
Z.AI GLM 4.6
Test
Dialogue tags
Scenario
Write 200 words with 90% dialogue
Run ID
2026-02-08-01KGZ07XN26VDCNTV4PXS377JQ
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
“The blueprints are a napkin, Bill, so focus.” “I *am* focused. It’s just… the cake has a volcano made of frosting. Lava.” “The primary objective is the gift mountain, not a sugar-based geological formation. The bouncy castle is the natural choke point, heavily guarded by sugar-crazed seven-year-olds.” “So I go in loud? Start a conga line?” “You will feign an ankle twist near the dip. That’s our window of opportunity during the ‘Pin the Tail on the Donkey’ phase.” “Right, right. And the dads? Any intel on their barbecue tongs? Are they the long-pronged variety?” “Unpredictable. Stick to the plan. When the music starts, we move.” “Just one more thing, Mol.” “What, Bill?” “Are we sure these are good presents? Last time, it was mostly bath bombs and novelty socks.”
created_at
"2026-02-08T16:10:43.129414+00:00"
Sunday, February 8th, 2026 at 4:10:43 PM GMT+00:00
model
"z-ai/glm-4.6"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
248
moderation_latency
(null)
generation_time
69572
tokens_prompt
61
tokens_completion
1736
native_tokens_prompt
66
native_tokens_completion
1752
native_tokens_completion_images
(null)
native_tokens_reasoning
1614
native_tokens_cached
32
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
0.0030656599999776
router
(null)
provider_responses
0
id
"chatcmpl-RH7ve71rCPz7ohfdEjfD9DCI"
status
200
is_byok
false
latency
54
endpoint_id
"c8607edc-8577-4199-b03d-1213da4743dc"
provider_name
"DeepInfra"
model_permaslug
"z-ai/glm-4.6"
api_type
"completions"
id
"gen-1770566973-dUxyezYOqh9dq7bndtiE"
upstream_id
"chatcmpl-RH7ve71rCPz7ohfdEjfD9DCI"
total_cost
0.0030656599999776
cache_discount
0.0000112000000224
upstream_inference_cost
0
provider_name
"DeepInfra"
Evaluation details
Result
Evaluator
Details
Meta Data
0.0000%
Matches word count
n/a
neededClean
false
words
135
36.7879%
Dialogue to Total Word Ratio
Ratio: 100.00%, Deviation: 10.00%
neededClean
false
wordsTotal
135
wordsDialogue
135
18.3940%