NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-24-01KQ032PT19RDGNVPHGKM1CEGQ
Run Details
Model:
Z.AI GLM 5.1
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
50.7s
Cost:
$0.024042
Provider:
Friendli
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
79.68%
Dialogue to Total Word Ratio
(details)
39.8425%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 70% 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 lowered the binoculars and rubbed her temples. The afternoon sun was brutal, but the sheer volume of high-value loot inside that suburban fortress made the sweat worthwhile. From the van, the target looked deceptively peaceful. Molly checked her watch. They had a three-minute window before the entertainer switched from juggling to magic, which would draw the crowd’s attention away from the gift table. She adjusted her earpiece and took a deep breath. "Is that a bouncy castle?" Bill asked, peering through the windshield with a look of childlike wonder. "It’s an inflatable containment zone, Bill. Keep up." "Right. Inflatable containment zone. So we breach the zone first?" "We don’t breach the zone. The zone is a diversion. We go around the petting zoo." "There’s a petting zoo?" Bill gasped. "Two goats and a profoundly depressed alpaca. It provides excellent cover for the flank maneuver. Stay focused." "What about the primary target?" Bill asked, squinting against the glare. "The presents, Bill. Focus. They’ll be stacked on the picnic table near the piñata." "I love piñatas." "You’re not here for candy. We grab the Lego sets, the Barbie dreamhouses, and whatever is in that giant sparkly gift bag. Then we exfiltrate to the van." "How do we get past the guards?" Bill whispered, suddenly ducking below the dashboard. "Guards? It’s a six-year-old’s birthday party, Bill." "I see a clown. Clowns are basically guards. They have a terrifying aura." "Clowns are terrifying, but they aren’t security. Just avoid eye contact and keep moving." "What if he makes a balloon animal for me?" "You decline politely! We are professionals. Act like it." "Fine. What’s the entry point?" "The side gate. It’s blocked by a luxury SUV stroller. We tiptoe past the sleeping baby in the carrier." "Sleeping baby? That’s high risk, Molly. They wake up, they scream, the whole op is blown. I saw a documentary about it once." "Which is why you don't step on a squeaky toy. Check your soles for pre-existing squeaks before we deploy." "My shoes are clean." "They have lights in the heels. They flash when you walk." "They're tactical illumination footwear!" "They light up like a disco ball, Bill! You'll give away our position!" "Fine, I'll walk on my hands." "Just wait for the sugar rush to hit. Once the cake is served, the targets will be too busy spinning in circles to notice us." "Then we grab the loot." "Then we grab the loot." "And the cake?" "Leave the cake." "But it looks like chocolate fudge. The good kind." "We are stealing thousands of dollars in high-end electronics and polymer bricks, and you want a slice of cake?" "I'm just saying, if the opportunity presents itself, I won't say no to a slice." "Grab the presents. Ignore the clown. Avoid the alpaca. Don't wake the baby. And no cake." "Got it. What if the piñata breaks while we're in there?" "You run, Bill. You run and don't look back." Bill nodded solemnly, reaching into the glove compartment to pull out a pair of black leather tactical gloves. He struggled to get the left one on, his knuckles cracking loudly as he did so. Molly sighed, pinching the bridge of her nose. It was going to be a very long afternoon.
created_at
"2026-04-24T15:52:33.356Z"
4/24/2026, 3:52:33 PM
model
"z-ai/glm-5.1-20260406"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
421
moderation_latency
(null)
generation_time
50402
tokens_prompt
77
tokens_completion
5875
native_tokens_prompt
66
native_tokens_completion
5443
native_tokens_completion_images
(null)
native_tokens_reasoning
5001
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"
usage
0.0240416
router
(null)
provider_responses
0
endpoint_id
"d4443f1c-fdb1-4cf0-a381-5b77b7e25ae7"
id
"chatcmpl-b042fcb840e94b7ebe1662d2302bf18d"
is_byok
false
latency
421
model_permaslug
"z-ai/glm-5.1-20260406"
provider_name
"Friendli"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.1; linux; x64))"
http_referer
(null)
request_id
"req-1777045953-vZDtGTjvmF3EYVeg3EG9"
session_id
(null)
api_type
"completions"
id
"gen-1777045953-4R6iDyyEtARa0Rjzhs7F"
upstream_id
"chatcmpl-b042fcb840e94b7ebe1662d2302bf18d"
total_cost
0.0240416
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Friendli"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
544
79.68%
Dialogue to Total Word Ratio
Ratio: 72.13%, Deviation: 2.13%
neededClean
false
wordsTotal
549
wordsDialogue
396
39.8425%