NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-02-01M1H9Z07TBZQBDFSKQ6FM6T6E
Run Details
Model:
Qwen 3.8 Max (Reasoning, XHigh)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
5m 30s
Cost:
$0.100802
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 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
Molly spread the birthday invitations across the van dashboard like stolen battle plans. The target was a backyard in Maple Court, crowded with balloons, sugar, and gifts. "The piñata is the weak point," Molly said. Bill blinked. "The piñata?" "Kids stare at piñatas. Adults stare at kids. Nobody watches gifts." Bill tapped his pencil. "What about clowns?" "There is no clown." "Always a clown." "No actual clowns?" Molly checked the invitation. "There is a magician. He arrives at two." "Then we arrive at one." They studied the crayon map. The garden had three entrances: gate, hedge gap, and bouncy castle. Bill took the hedge. Molly took toddlers. "I am not doing toddler duty again," Bill said. "You have gentle hands." "I gave a child a balloon and he proposed." "That is your reputation." Bill sighed. "What is the signal?" "When I sneeze twice, grab presents." "And if it is allergies?" "Then grab presents and antihistamine." Molly had prepared for every likely obstacle, including fence locks, enthusiastic uncles, and a possible karaoke machine. She had timed the karaoke risk at four minutes, longer if a grandmother remembered ABBA. Bill carried a coil of rope, though he could not tie anything except a shoelace, and even that was negotiable. Their escape vehicle was a silver hatchback with a dent shaped like a duck, which Molly refused to explain. The driveway would be blocked by a clown car, if there was a clown, and a bouncy castle, if there was not. She had drawn red arrows on the map, then erased them because red attracted attention, she said, though Bill suspected she simply disliked the color of ambition. She had also rehearsed a cover story. Molly was an assistant to the deputy coordinator of festive logistics. Bill was her silent intern, which suited him because he forgot words when children looked at him with hope. If questioned, Molly would speak in acronyms until the adult wandered away. If trapped, Bill would offer the banana. It was, he insisted, fruit. The plan needed timing. Molly would enter as a cupcake helper. Bill would hide by the recycling bins, which made him feel judged. He would sneeze once to confirm survival. "Why are we stealing from children?" Bill asked. Molly frowned. "We are stealing from insurers, through children." "That sounds worse." "It sounds professional." The magician worried them. Magicians noticed things, especially missing doves. Molly promised no magic would be recruited unless useful. Bill packed sacks, a wagon, a dog whistle, and a banana. "Emergency," Bill said. "Define emergency." "A monkey." "There will be no monkey." "There is always a monkey." "There is a magician." "Same family." Molly folded the map. "No heroics. We take gifts, leave piñata, and avoid bouncy castle." "I am professional." "You bounced for reconnaissance." "It was research." She started the van. Tomorrow Maple Court would learn the real party danger was not sugar, magicians, or piñatas. It was two thieves with crayon maps, bananas, and no respect for wrapping paper.
created_at
"2026-09-02T14:56:55.559Z"
9/2/2026, 2:56:55 PM
model
"qwen/qwen3.8-max-20260803"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1870
moderation_latency
(null)
generation_time
330245
tokens_prompt
75
tokens_completion
12942
native_tokens_prompt
127
native_tokens_completion
16758
native_tokens_completion_images
(null)
native_tokens_reasoning
16051
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
0.100802
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788361015-lKv89KRd3N47VH3fQZ4i"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788361015-wJ3tbsfnZQTWLAo3aSzG"
upstream_id
"chatcmpl-795602cc-a6b1-932f-aaec-a93f340e1496"
provider_responses
0
endpoint_id
"6332e9fd-e0c4-4cae-8c17-d25bc774864a"
id
"chatcmpl-795602cc-a6b1-932f-aaec-a93f340e1496"
is_byok
false
latency
1870
model_permaslug
"qwen/qwen3.8-max-20260803"
provider_name
"Alibaba"
status
200
total_cost
0.100802
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: 30.00%, Deviation: 0.00%
neededClean
false
wordsTotal
500
wordsDialogue
150
100.0000%