NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-02-26-01KJBKZ5PBT85E2E8XJSXEWZFY
Run Details
Model:
Qwen 3.5 122B
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
1m 13s
Cost:
$0.020981
Provider:
Alibaba
Evaluation summary
Score
Evaluator
Details Link
99.74%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
49.8722%
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 tapped the flyer on the dashboard. "Saturday. Three PM. The Johnson residence." Bill squinted at the paper. "That's a birthday party, Molly. For a six-year-old." "Exactly. Maximum chaos, minimum security." "The security is the parents, Molly. They're armed with cake and judgment." "I'm talking about the presents. Under the tree. Wrapped in foil." Bill rubbed his temples. "We're stealing from kids?" "We're redistributing wealth. From the rich kid to the professional thieves." "That's not how wealth redistribution works." "It is in our line of work. What's the gift list?" Bill scanned the paper. "A bike. A tablet. A hamster." "A hamster?" "A hamster named Sir Fluffs." "We can't fence a hamster." "We keep the hamster. It's morale." "Molly, we are not keeping a rodent in the getaway van." "Why not? It's low maintenance." "It poops. It squeaks. It attracts cops." "Fine. Leave the hamster. Focus on the bike." "The bike is locked to the porch." "Cut the chain. Use the bolt cutters." "The parents will hear the cut." "Then we cut silently. Like ninjas." "Ninjas don't steal from six-year-olds." "Ninjas steal from anyone who underestimates them." Bill sighed. "Okay. What about the cake?" "Leave the cake. Too sticky." "What about the piƱata?" "That's a distraction. We smash it, they run, we grab the loot." "We're breaking into a party, Molly. Not a bank." "Same difference. Parents are the guards. Kids are the civilians." "Civilians who scream." "High-pitched screams. Hard to track." "I'm worried about the balloon animals." "Why?" "They pop. Loudly." "We wear earplugs." "Earplugs make us look suspicious." "We wear clown noses. Blend in." "We're not clowns." Bill looked at her. "You're insane." "I'm efficient. You're the muscle. I'm the brain." "I'm the guy who carries the bike." "Exactly. Can you fit a bike in the trunk?" "It's a hatchback." "Then we leave the bike. Take the tablet." "The tablet is password protected." "We'll find the password. Probably 'password'." "Or 'ilovecats'." "We need a plan for the dog." "The dog is a chihuahua." "It yaps." "We bribe it with a treat." "We don't have treats." "We steal a treat from the party." "Stealing from a party to steal from a party." "Circular economy." Bill nodded slowly. "Alright. Saturday. Three PM. No hamster." "No hamster." "And no clown noses." "Fine. Just masks." "Black masks." "Black masks. Like bank robbers." "We aren't bank robbers." "We're birthday robbers." "That sounds worse." "Let's go over the exit strategy." "Back door. Through the laundry room." "There's a washing machine." "Perfect. We hide inside." "You're joking." "I'm not. Spin cycle hides sound." "Molly, we will be in a dryer." "Washer. Spin cycle." "We'll get dizzy." "We'll be dizzy and rich." "You'll be dizzy." "And broke." "No." "Rich." "I hate this plan." "You love the money." "I love the money." "Then shut up and pack the bags." "I'm packing." "Good. Grab the hamster just in case." "Molly!" "Kidding. Mostly." "Don't tell the kids." "They'll never know." "They'll cry." "Let them cry."
created_at
"2026-02-26T00:00:52.945Z"
2/26/2026, 12:00:52 AM
model
"qwen/qwen3.5-122b-a10b-20260224"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
446
moderation_latency
(null)
generation_time
72798
tokens_prompt
77
tokens_completion
6447
native_tokens_prompt
76
native_tokens_completion
6547
native_tokens_completion_images
(null)
native_tokens_reasoning
5678
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
0.0209808
router
(null)
provider_responses
0
endpoint_id
"33d466d1-d2f7-45ed-8ff3-fda7d2e32351"
id
"chatcmpl-0a65507e-af94-93a0-8066-8324a4eecacb"
is_byok
false
latency
446
model_permaslug
"qwen/qwen3.5-122b-a10b-20260224"
provider_name
"Alibaba"
status
200
api_type
"completions"
id
"gen-1772064052-Rrs29yMPekDRfOMbTfPX"
upstream_id
"chatcmpl-0a65507e-af94-93a0-8066-8324a4eecacb"
total_cost
0.0209808
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Alibaba"
Evaluation details
Result
Evaluator
Details
Meta Data
99.74%
Matches word count
n/a
neededClean
false
words
496
0.00%
Dialogue to Total Word Ratio
Ratio: 94.21%, Deviation: 24.21%
neededClean
false
wordsTotal
501
wordsDialogue
472
49.8722%