NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-02-25-01KJBFNR5KCY13089E1NDXSCP4
Run Details
Model:
Qwen 3.5 35B
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
51.2s
Cost:
$0.016061
Provider:
Alibaba
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0000%
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 adjusted the latex ear on her head. The house was a pristine suburban brick structure with a manicured lawn. "Are you absolutely sure this is the location?" Bill peered through the sheer lace curtain. "According to the flyer, yes. But I smell popcorn." "Perfect. Popcorn means they are settled. Settle means presents are unboxed." "We want presents?" "Yes. The wrapping paper contains the value. The toys inside are secondary." "Why not just cash?" "Cash is tracked. Toys are anonymous. A plastic race car is just a plastic race car. No serial number." Bill shifted his weight uncomfortably against the window frame. "This feels wrong. I thought we were doing banks." "Banks have lasers. Kids have cake." "Cake is sticky." "Cake is the perfect cover. While the mom blows out the candles, we move in. We sweep." "Sweep?" "We take the boxes. The pile goes into these bags." She pointed to black garbage sacks. "Garbage bags?" "Stealth sacks. Opaque to the eye. No logos." "What if the dad catches us?" "Then we pretend to be the hired entertainment. We say we are the fire dancers." "We are not fire dancers." "We will be. We will hold burning torches made of marshmallows." "That is not safe." "It is comedy. They won't call the cops on a dad protecting his kids from fire dancers." Bill looked at the window. "What if they have security cameras?" "We wear sunglasses. Inside. They won't notice. It's a kids' party. Kids don't understand sunglasses." "I don't understand why we need sunglasses." "Psychological warfare." "I hate psychological warfare." "Stop whining. The doorbell just rang." "Who is it?" "The guests. We need to get in." "How?" "Knock. Then we run when the door opens." "I hate running." "Running is part of the getaway plan." "The plan has no exit strategy." "We exit through the window." "The window is on the second floor." "There is a tree. We will use the tree." "That tree is dead." "Then we will use the hedge." "I do not want to climb a dead tree with my pants full of toys." "Trust me. I am the master of this operation." "You are the master of losing." "Focus. When the singing starts, we enter." "Why singing?" "It drowns out the sound of us stealing." "It drowns out the sound of us crying." "Crying? Why would we cry?" "Because the tree is dead." "Okay. Okay. Let's go." "Wait, one more thing." "What?" "Do we take the cake?" "Absolutely not. It is evidence." "It tastes like butter." "We take the gifts. Leave the food." "Fine." "Ready?" "No." "Then we are not ready." "Okay. Ready." "Let's go." "Time is ticking." "What time is it?" "Ten to." "How long until singing?" "Five minutes." "Five minutes?" "That is enough time." "No it is not." "Yes it is." "I have a bad feeling." "Don't feel." "I am feeling." "Stop feeling." "I am afraid." "Fear is useful." "Fear is bad." "Fear makes us fast." "I want to be slow." "We are not slow." "We are clumsy." "We are agile." "I am not agile." "You will be." "I don't think so." "Think later. Act now." "Action is scary." "Inaction is boring." "I want boring." "You want money." "I want safety." "Money is safety." "Money is trouble." "Trust me." "I do not." "Trust it." "Do we have gloves?" "Yes." "They are latex?" "Yes." "Do they fit?" "Yes." "Good." "What if the dog barks?" "We bribe the dog." "With what?" "With meat." "What meat?" "The chicken from the party." "We are stealing chicken too?" "Yes." "Okay. We go." "Okay. We go."
created_at
"2026-02-25T22:45:49.883Z"
2/25/2026, 10:45:49 PM
model
"qwen/qwen3.5-35b-a3b-20260224"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
402
moderation_latency
(null)
generation_time
51002
tokens_prompt
77
tokens_completion
8010
native_tokens_prompt
76
native_tokens_completion
8021
native_tokens_completion_images
(null)
native_tokens_reasoning
7065
native_tokens_cached
0
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
0.016061
router
(null)
provider_responses
0
endpoint_id
"48c1210f-b7d2-4688-a340-f7424fc99d2b"
id
"chatcmpl-fc4997d1-fb8e-9bba-8b60-3f6ec5e5d9f2"
is_byok
false
latency
402
model_permaslug
"qwen/qwen3.5-35b-a3b-20260224"
provider_name
"Alibaba"
status
200
api_type
"completions"
id
"gen-1772059549-Wg4GQWwjvTKRxwLuU478"
upstream_id
"chatcmpl-fc4997d1-fb8e-9bba-8b60-3f6ec5e5d9f2"
total_cost
0.016061
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Alibaba"
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
597
0.00%
Dialogue to Total Word Ratio
Ratio: 92.13%, Deviation: 22.13%
neededClean
false
wordsTotal
597
wordsDialogue
550
0.0000%