NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-27-01KQ6Z9AH0E6ETD52SCX6YW906
Run Details
Model:
Qwen 3.5 Plus (2026-04-20)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
2m 25s
Cost:
$0.023145
Provider:
Alibaba
Evaluation summary
Score
Evaluator
Details Link
99.98%
Matches word count
(details)
92.05%
Dialogue to Total Word Ratio
(details)
96.0172%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 50% 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 spread the crayon blueprint on the damp basement table, right over a grease-stained map of the subway system. Bill picked at the sequins glued to his tactical vest, his jaw tense beneath the fraying wool of his balaclava. "Henderson residence," Molly hissed, tapping a waxy circle drawn in purple marker. "Pirate theme. Intel confirms one PS5 and enough gift cards to fund our retirement. We hit at three o'clock sharp." Bill groaned, tugging the oversized plastic hook over his leather glove. "A PS5? At a seventh birthday? Molly, are you sure the parents aren't just trolling the invite list?" "The cake is a distraction. We grab the loot while they slice." Molly grabbed a plastic cutlass from the table and jabbed it at the floor. "Infiltration is key. We go in as the entertainment." "I knew it." Bill kicked a stacked pile of unmarked cardboard boxes, sending loose marbles skittering across the concrete. "The clown routine? I squeaked so loud the driver pulled over last time." "Focus, Bill. Clowns are invisible in plain sight. No parent questions a harlequin until the toys vanish. You create the chaos. Balloon animals, juggling the piñata. I load the sacks." "Juggling the piñata? That's fifty pounds of papier-mâché and hard candy. I'm a thief, not a circus act with a hernia." "You'll manage. We exit through the garden shed. No alarms, no witnesses, just the sound of weeping children and our escaping tires." Molly grabbed two burlap sacks embroidered with happy skulls and stuffed duct tape into her pocket. "Suit up. If we get caught, we blame the magician." Molly applied a stripe of white greasepaint to her cheek. She visualized the layout. The living room contained a mountain of wrapped boxes near the fireplace, guarded by a sleeping golden retriever and a platter of finger sandwiches. Bill fumbled with the buckles on his jousting doublet, his fingers thick and clumsy. "What about the drone?" Bill pointed a gloved finger at the map. "Timmy's uncle is bringing the DJI Mavic. If we move too fast, he'll broadcast our exit to the neighborhood watch chat." "The drone battery is a dud. I swapped it during the grocery run. We have a fifteen-minute window before he notices." Molly grabbed the squeaky shoes and shoved them into Bill's chest. "Move fast, Bill." Bill sighed, staring at the red leather with hatred. "Fine. But if I trip over a stray balloon, I'm blaming you. And we're taking the Lego set first. That's the portable value." Molly shoved the red foam nose onto Bill's face. "Deal. And check the masks. Mrs. Higgins monitors the scanners; she'll recognize my voice if I squeak." The damp air hung heavy with the scent of damp wool and desperation. A single lightbulb swung overhead, casting long shadows that danced like goblins across the concrete walls. Bill's stomach growled, a reminder that the last meal they'd shared was a stolen bag of pretzels from three days ago. "Now put on the nose. We're burning daylight."
created_at
"2026-04-27T08:00:56.871Z"
4/27/2026, 8:00:56 AM
model
"qwen/qwen3.5-plus-20260420"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
703
moderation_latency
(null)
generation_time
145155
tokens_prompt
77
tokens_completion
9419
native_tokens_prompt
76
native_tokens_completion
9631
native_tokens_completion_images
(null)
native_tokens_reasoning
8931
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
0.0231448
router
(null)
provider_responses
0
endpoint_id
"8b4a0fbb-6705-4f8d-a7f1-ad18ecc0edbc"
id
"chatcmpl-f3bf133a-8fae-99fa-894c-c175ddedd759"
is_byok
false
latency
703
model_permaslug
"qwen/qwen3.5-plus-20260420"
provider_name
"Alibaba"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.1; linux; x64))"
http_referer
(null)
request_id
"req-1777276856-ncn4oNW9tsqlcXJ8KxoR"
session_id
(null)
api_type
"completions"
id
"gen-1777276856-5QogHyR7BsiOzaCLUXdq"
upstream_id
"chatcmpl-f3bf133a-8fae-99fa-894c-c175ddedd759"
total_cost
0.0231448
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Alibaba"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
99.98%
Matches word count
n/a
neededClean
false
words
502
92.05%
Dialogue to Total Word Ratio
Ratio: 51.29%, Deviation: 1.29%
neededClean
false
wordsTotal
505
wordsDialogue
259
96.0172%