NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-02-01M1H9QT25TYDVBT4JEGT12HKP
Run Details
Model:
Qwen 3.8 Max (Reasoning, XHigh)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
4m 47s
Cost:
$0.087644
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 spread the crayon map across the van floor while Bill polished a tiny wand. “The party starts at two. We enter as entertainers and leave with every present.” “Are you suggesting we become clowns? I refuse to become a clown for children.” “You already own clown shoes. They honk when you step on the brake pedal.” “Those are orthopedic shoes. The honking is an emotional condition, not a criminal weakness.” Bill frowned at the map while a rubber duck slid between them like a witness. “Fine. What is our cover? A wizard? A pirate? A disappointed birthday party inspector?” “You are a wizard. You have a goatee, and wizard robes hide stolen gifts.” “That only works if I carry a staff and mutter nonsense about birthday magic.” “Perfect. The magic words are please hand over the presents, followed by a wink.” “That is not magic. That is theft wearing a hat, and hats attract attention.” “Attention is fine. Parents ignore entertainers until someone touches the cake or the piñata.” Molly drew a circle around the backyard and labeled it Presents with three crayon strokes. “We place the gifts in donation bags. Nobody questions donation bags at a party.” “What if a child cries? Small children have loud mouths and sticky investigative fingers.” “You cry louder. Children respect volume, and volume is basically a plan in disguise.” “That is not comforting. I prefer plans without crying, unless the crying is mine.” Bill picked up the rubber duck and examined it like a tiny yellow legal expert. “The duck is emotional support. First for me, then for the tiny screaming hostages.” “They are not hostages. They are guests, and guests expect cake, games, and refunds.” “Then we provide games. The game is called Find the Presents Before Daddy Notices.” “That game has too many words, and Daddy sounds suspiciously like a security guard.” “Daddies are easy. You ask if the grill is off, and they vanish quickly.” Molly placed a party hat on the dashboard like a general accepting a ridiculous promotion. “My signal is this hat. You grab the presents while I coordinate imaginary cupcakes.” “Imaginary cupcakes? That sounds like a crime that begins with disappointment and ends badly.” “Crime usually ends badly. We simply choose the ending where we own a van.” “We already own a van. It smells like old candy and poor decisions, obviously.” “Exactly. The van is character. Character explains many stains and several missing license plates.” “We have license plates? I thought those were optional, like brakes or honest work.” Bill tucked the duck into his coat pocket while the wind carried distant balloon music. Molly checked her watch and smiled like someone who had mistaken chaos for a calendar. Molly handed him a kazoo and a grin shaped like a tiny tired birthday unicorn. Somewhere beyond the hedge, a child screamed with joy over a poorly shaped balloon nearby. The presents waited in the yard like small brightly wrapped mistakes begging to be rescued.
created_at
"2026-09-02T14:52:59.853Z"
9/2/2026, 2:52:59 PM
model
"qwen/qwen3.8-max-20260803"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
781
moderation_latency
(null)
generation_time
287223
tokens_prompt
75
tokens_completion
12009
native_tokens_prompt
127
native_tokens_completion
14565
native_tokens_completion_images
(null)
native_tokens_reasoning
13916
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
0.087644
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788360779-00dxBbzgx9dpJWvLldje"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788360779-I8yWUhFT26PPM92qFBHx"
upstream_id
"chatcmpl-60f75e97-1b40-9ee4-8a5e-768fe30b054f"
provider_responses
0
endpoint_id
"6332e9fd-e0c4-4cae-8c17-d25bc774864a"
id
"chatcmpl-60f75e97-1b40-9ee4-8a5e-768fe30b054f"
is_byok
false
latency
781
model_permaslug
"qwen/qwen3.8-max-20260803"
provider_name
"Alibaba"
status
200
total_cost
0.087644
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%