NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-02-25-01KJBH1P8YQZ54D43KP90RPAQE
Run Details
Model:
Qwen 3.5 27B
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
56.1s
Cost:
$0.013861
Provider:
Alibaba
Evaluation summary
Score
Evaluator
Details Link
1.04%
Matches word count
(details)
2.81%
Dialogue to Total Word Ratio
(details)
1.9234%
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 stared at the blueprints spread across the dashboard of the van. She tapped a finger on the drawing of the suburban house. The paper crinkled under her nail. She looked at the clock on the wall. The engine hummed quietly in the background. "You're serious about this?" she asked. "It's a birthday party for a six-year-old." Bill adjusted his black ski mask, though they were in a parking lot. He chewed on a piece of gum. He pointed at the list with a gloved hand. "The risk is low, the reward is high. Look at the manifest." "It's a list of Lego sets and a tricycle." "High-end Lego. Limited edition Star Wars. That tricycle has gold plating." "It's a plastic bike, Bill." "It's a statement piece. We take the whole haul. Every box, every balloon." "The parents are inside. The security is... a dad with a cake." "Exactly. No laser grids. No motion sensors. Just cake." "And if we get caught?" "We say we're the gift delivery service. Who checks ID at a birthday?" "We're thieves, not couriers." "Details. We need the loot bag ready. The one with the silent lining." "Why silent?" "So the wrapping paper doesn't crinkle when we run." "You're overthinking the noise of paper." "I'm overthinking everything. That's why we're good." "We're good because we're desperate. We need cash, not a Tonka truck." "Fences pay top dollar for unopened toys. Kids trade them like stocks." "This is insane." "It's art. Now, check the timing." "Party starts at two. Cake at two-thirty." "Perfect window. We go in during the song." "Happy Birthday?" "The distraction is the singing. We slip in the back door." "The back door has a lock." "I have a lockpick set in my pocket. It's shaped like a lollipop." "You brought a lollipop pick?" "It's thematic." "You're ridiculous." "I'm profitable. Now, do we have the van ready?" "It's a minivan. It says 'Mom' on the bumper." "Perfect cover. We look like we belong." "We look like we're picking up kids from soccer." "Same thing. Chaos and noise." "Okay. I'll get the bags." "Don't forget the glitter bombs." "What?" "If they chase us, we blind them with confetti." "That's not a distraction, that's a mess." "It's a tactical mess. Now, move it." Molly grabbed the bags. She looked at Bill. The sun was setting behind them. The engine hummed quietly in the background. She checked the zipper on the bag. "What about the clown?" "We ignore the clown. Clowns are easy to bribe with cake." "You think a clown will eat cake?" "He's a professional. He eats everything." "And the balloons?" "Leave the balloons. They are heavy." "No, we take them. They look like loot." "Fine. But I'm not holding a balloon." "If we get arrested, I'm telling them you planned it." "Fair enough. Let's get those presents."
created_at
"2026-02-25T23:09:49.73Z"
2/25/2026, 11:09:49 PM
model
"qwen/qwen3.5-27b-20260224"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
519
moderation_latency
(null)
generation_time
55991
tokens_prompt
77
tokens_completion
5855
native_tokens_prompt
76
native_tokens_completion
5766
native_tokens_completion_images
(null)
native_tokens_reasoning
5048
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
0.0138612
router
(null)
provider_responses
0
endpoint_id
"f3f88554-0c22-4d57-973b-0dec948808e8"
id
"chatcmpl-fb8cf336-cd8f-93f8-ae48-1f9584cb26d3"
is_byok
false
latency
517
model_permaslug
"qwen/qwen3.5-27b-20260224"
provider_name
"Alibaba"
status
200
api_type
"completions"
id
"gen-1772060989-thumCvmBUmMf8hdzxxG3"
upstream_id
"chatcmpl-fb8cf336-cd8f-93f8-ae48-1f9584cb26d3"
total_cost
0.0138612
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Alibaba"
Evaluation details
Result
Evaluator
Details
Meta Data
1.04%
Matches word count
n/a
neededClean
false
words
474
2.81%
Dialogue to Total Word Ratio
Ratio: 78.45%, Deviation: 8.45%
neededClean
false
wordsTotal
478
wordsDialogue
375
1.9234%