NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-02-01M1HC250NJ9Q4392TB4YHG7J3
Run Details
Model:
Qwen 3.8 Max (Reasoning, XHigh)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
5m 10s
Cost:
$0.079742
Provider:
Alibaba
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
100.00%
Dialogue to Total Word Ratio
(details)
100.0000%
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 unrolled a crayon blueprint across the dashboard of the van. Bill adjusted his tiny black mask and looked deeply worried. "The target is a children's party with too many presents." "How many children?" "Twenty. Small, loud, and armed with frosting." "Frosting is a serious tactical hazard." "Exactly. Bring napkins." Bill scribbled napkins on his list beside a doodle of cake. "What is our cover?" "You are the magician. I am the extremely boring uncle." "I hate children." "Then hate quietly." "Where are the presents?" "Under a table shaped like a pirate ship." "Kids guard pirate ships." "Kids also guard cake." "We distract them with cake." "With cake?" "Yes. Cake is a delicious smoke bomb." Molly demonstrated by stuffing a cupcake into a sock. "That is not a smoke bomb." "It becomes one in the right hands." "Whose hands?" "Mine, obviously." Outside the fence, balloons bounced like suspicious colored planets. "What about the parents?" "Parents believe anything if you hold a clipboard." "I have a clipboard." "Why do you have a clipboard?" "I stole it from a school." "Perfect. School things are criminal." Bill practiced stern parental nods while Molly checked her candy pockets. "Entry route?" "Through the hedge behind the bouncy castle." "Hedges scratch." "Scratches build character." "I already have character." "Then you can afford more." A child screamed beyond the fence, probably over a blue balloon. "What if someone cries?" "We compliment their shoes." "And if they still cry?" "Deploy glitter?" "No glitter. Glitter is forever." "Forever is bad for escapes." Molly drew an arrow through a drawing of a piñata. "Exit through the piñata?" "No. Exit beside it, carrying gift bags." "Gift bags are bulky." "Bulky means generous." "Generous means caught." "Then we are not generous today." Bill tucked a small paper crown into his jacket for disguise. "What is the signal?" "I sneeze twice." "And if you only sneeze once?" "Then I am enjoying the party." "We are supposed to steal." "Stealing can be enjoyable." Molly tested a tiny grappling hook made from a coat hanger. "Is that necessary?" "Everything is necessary." "Even the balloon?" "Especially the balloon." "It might pop." "Then we pop with dignity." Bill sighed and folded a napkin into a tiny burglar mask. "What about the dog?" "Cupcake has been handled." "Handled how?" "With cheese." "Good. Cheese is diplomacy." "You are learning." A banner flapped overhead, reading HAPPY BIRTHDAY in aggressive purple letters. "Final plan?" "Enter during cake." "Bag all presents." "Smile at parents." "Avoid glitter." "Exit with dignity." "And balloon?" "Balloon optional." "Balloon essential." "Fine. Balloon essential." Molly zipped her jacket and winked at the wobbly party gate. "Ready?" "No." "That is the spirit." "Let us steal from children." "We are terrible people." "Terrible people get presents." Bill opened the van door, and birthday music attacked them immediately. "After you." "Together?" "Together." Bill nodded. "Remember, if caught, we are lost clowns." "We are not clowns." "You are wearing a red nose." "It helps me breathe." "Then breathe quietly." "Impossible." "Quietly." "Fine."
created_at
"2026-09-02T15:33:35.902Z"
9/2/2026, 3:33:35 PM
model
"qwen/qwen3.8-max-20260803"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1207
moderation_latency
(null)
generation_time
310143
tokens_prompt
75
tokens_completion
10559
native_tokens_prompt
127
native_tokens_completion
13248
native_tokens_completion_images
(null)
native_tokens_reasoning
12418
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"
service_tier
(null)
usage
0.079742
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788363215-JBPeZ2URPPdenFqAO9X9"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788363215-E1LRRUk3MnjAWXmOnUEc"
upstream_id
"chatcmpl-fa59abf0-f04b-9944-9a8a-0e5e519c52c2"
provider_responses
0
endpoint_id
"6332e9fd-e0c4-4cae-8c17-d25bc774864a"
id
"chatcmpl-fa59abf0-f04b-9944-9a8a-0e5e519c52c2"
is_byok
false
latency
1206
model_permaslug
"qwen/qwen3.8-max-20260803"
provider_name
"Alibaba"
status
200
total_cost
0.079742
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Alibaba"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
500
100.00%
Dialogue to Total Word Ratio
Ratio: 70.00%, Deviation: 0.00%
neededClean
false
wordsTotal
500
wordsDialogue
350
100.0000%