NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHK20MPR9B1EA2W1HA2Z2A4X
Run Details
Model:
Qwen 3.5 Plus (2026-02-15)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
37.3s
Cost:
$0.001764
Provider:
Alibaba
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 30% 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 rain slicked the neon plastic letters spelling out HAPPY BIRTHDAY on the front lawn, making them look like a failed science experiment in glow-in-the-dark slime. Molly crouched behind a inflatable T-Rex that had lost half its air, watching the suburban house buzz with the chaotic energy of six-year-olds. Inside, the shrieks of children playing musical chairs sounded less like fun and more like a siren warning of impending doom. Bill adjusted his fake mustache, which was currently sliding off his lip due to the humidity, and checked his watch for the third time in ten seconds. "We are doing this," Molly whispered, her eyes locked on the mountain of wrapped boxes under the tree. "Tonight, we become legends." Bill tugged at the collar of his ill-fitting clown suit. "Legends? Molly, we are dressed as expired circus performers. If we get caught, the headlines will read 'Clowns Nabbed Stealing Lego Sets.' It's humiliating." "It's genius," she countered, crawling forward through the wet grass. "Nobody suspects the clown. Nobody ever looks twice at the guy holding a balloon animal sword. We walk in, we grab the loot, we walk out before they even cut the cake." "The cake is probably dry anyway," Bill muttered, trying to untangle a string of party beads from his oversized shoes. "And that T-Rex is judging me. I can feel its plastic eyes burning into my soul." "Focus, Bill. Think of the score. Those aren't just toys. That is high-end electronics disguised as plastic garbage. Tablets, gaming consoles, enough cashmere sweaters for toddlers to fund our retirement in Belize." "Belize doesn't take payment in Paw Patrol figurines, Molly." "They take the credit cards inside the gift bags, you pessimist." They reached the side door, where a stack of discarded pizza boxes provided just enough cover. Molly pulled a set of lockpicks from her pocket, but then paused, realizing the door was slightly ajar. A stream of golden light and the smell of burnt hot dogs spilled out onto the porch. "Open door," Bill noted, his voice trembling slightly. "That's either a trap or terrible parenting." "It's an invitation," Molly said, pushing the door open just wide enough to slip through. "Stay low. Stick to the shadows. And for the love of god, do not honk your nose." "I make no promises regarding the nose," Bill whispered back as they entered the hallway. The living room was a war zone of streamers and crushed paper plates. The mountain of gifts loomed ahead, a glittering fortress of consumerism. Molly signaled for Bill to start stuffing large black trash bags with the biggest boxes. They moved with surprising efficiency, swiping a massive dollhouse and a stack of video games. Just as Bill reached for the final prize, a giant teddy bear wearing a party hat, the music stopped. The room went silent. From the kitchen, a small voice echoed into the living room. "Mommy, the clowns are stealing my presents." Molly froze, a bag full of loot slung over her shoulder. Bill slowly turned, his fake mustache now hanging by a single thread of glue. "Run," Molly said. "Gladly," Bill replied, and they bolted toward the back door, tripping over a tricycle in their haste to escape the wrath of a dozen sugar-crazed children.
created_at
"2026-02-16T11:05:58.590085+00:00"
2/16/2026, 11:05:58 AM
model
"qwen/qwen3.5-plus-20260216"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1958
moderation_latency
(null)
generation_time
37262
tokens_prompt
77
tokens_completion
837
native_tokens_prompt
78
native_tokens_completion
722
native_tokens_completion_images
(null)
native_tokens_reasoning
0
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
0.001764
router
(null)
provider_responses
0
id
"chatcmpl-691b1234-fc08-9e6d-8bc2-47f78babecd5"
status
200
is_byok
false
latency
1958
endpoint_id
"71e0f6c3-7815-4734-92d3-8a04f3b0ca91"
provider_name
"Alibaba"
model_permaslug
"qwen/qwen3.5-plus-20260216"
api_type
"completions"
id
"gen-1771239920-ilGgZeO0414T53Li5EN5"
upstream_id
"chatcmpl-691b1234-fc08-9e6d-8bc2-47f78babecd5"
total_cost
0.001764
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Alibaba"
Evaluation details
Result
Evaluator
Details
Meta Data
0.0000%
Matches word count
n/a
neededClean
false
words
553
51.6562%
Dialogue to Total Word Ratio
Ratio: 33.63%, Deviation: 3.63%
neededClean
false
wordsTotal
553
wordsDialogue
186
25.8281%