NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-02-01M1HB62QP1SCT82FE7DKV2HJ3
Run Details
Model:
Qwen 3.8 Max (Reasoning, XHigh)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
4m 59s
Cost:
$0.082358
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 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 map across the dirty van floor. "The birthday fortress is actually just a bounce house, Bill." Bill then checked his disguise in the cracked rearview mirror. "You say fortress, I say bouncy death trap for kids." Molly ignored him and circled a purple balloon with crayon. "The clown balloon guards the presents, so we avoid it." Bill shuddered because clowns once chased him through a mall. "Air can be bribed if we bring a pump today." Molly slapped a laundry basket onto the map with authority. "No pumps. We carry baskets and look exhausted near cake." Bill practiced an exhausted parent stare in the mirror again. "Exhausted parents never smile. They only stare into cake slowly." Molly drew tiny mustaches over every adult face on paper. "You wear three mustaches, because extra mustache means authority today." Bill touched his upper lip proudly, revealing two crooked mustaches. "What about the pinata? It has angry paper eyes too." Molly crossed the pinata from her list with red pen. "Ignore the pinata unless it moves, which is very unusual." Bill remembered a donkey pinata swinging at his head once. "They always move when scared, and I am scared now." Molly patted Bill on the shoulder with a gloved hand. "Then the pinata will respect you, maybe even salute you." Bill considered this while stacking candy bars into his pockets. "You promise that no child will cry today, Molly, right?" Molly hesitated because promises near children were very legally difficult. "We steal presents, not souls, so some crying remains possible." Bill nodded slowly, accepting the moral limits of present theft. "Define minimal crying before I put on the fourth mustache." Molly held up one gloved finger like a tiny professor. "One shared cry. Everyone cries briefly, then eats cake quickly." Bill seemed comforted by mathematical sadness and folded the map. "I can work with one cry, if it is shared." Molly rolled the map into a neat pink criminal scroll. "Operation Sticky Fingers begins at three, because cake arrives then." Bill checked his watch, then licked frosting from his thumb. "Why are our fingers sticky before the birthday party starts?" Molly stared at Bill until he stopped licking his thumb. "Because children always touch cake, then presents, then everything nearby." Bill imagined tiny fingerprints covering his favorite black burglar gloves. "Maybe we need tiny gloves for tiny criminal hands too." Molly handed Bill a packet of wipes from her pocket. "Wipe presents first, then always steal them with clean hands." Bill tucked the wipes beside a rubber duck and rope. "What if the birthday kid sees us carrying loot away?" Molly produced two gift bags decorated with green glittery dinosaurs. "We place stolen presents inside boring adult tax documents today." Bill admired the plan despite fearing balloons, children, and pinatas. "Tax documents defeat suspicion, because no child reads taxes ever." Molly opened the van door as distant party music began. "Remember, Bill, we are only stealing joy politely today, mostly."
created_at
"2026-09-02T15:18:16.065Z"
9/2/2026, 3:18:16 PM
model
"qwen/qwen3.8-max-20260803"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1192
moderation_latency
(null)
generation_time
298775
tokens_prompt
75
tokens_completion
12260
native_tokens_prompt
127
native_tokens_completion
13684
native_tokens_completion_images
(null)
native_tokens_reasoning
13006
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
0.082358
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788362296-CCG5R6gH4arR03ReHOrk"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788362296-KDW5hAfcrp0YPVs3HXHF"
upstream_id
"chatcmpl-58137ede-3295-9de6-9ab3-f4486d70ed59"
provider_responses
0
endpoint_id
"6332e9fd-e0c4-4cae-8c17-d25bc774864a"
id
"chatcmpl-58137ede-3295-9de6-9ab3-f4486d70ed59"
is_byok
false
latency
1192
model_permaslug
"qwen/qwen3.8-max-20260803"
provider_name
"Alibaba"
status
200
total_cost
0.082358
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: 50.00%, Deviation: 0.00%
neededClean
false
wordsTotal
500
wordsDialogue
250
100.0000%