NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-02-01M1HH2F82J433DTVPG4TSNWRQ
Run Details
Model:
Qwen 3.8 Flash (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
3m 10s
Cost:
$0.010461
Provider:
Alibaba
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
98.79%
Dialogue to Total Word Ratio
(details)
99.3958%
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 crouched behind the recycling bin, checking her watch. Bill arrived in a tuxedo and inflatable arm floaties. "Do you need the floaties?" Bill asked. "It's called stealth," Molly said. "We're in a backyard." "Exactly. No witnesses but cake." Bill pointed at the gift table. "There are thirty presents." "Twenty-nine. One has a cursed yo-yo." "Why?" "His mother is chaotic." Bill pulled a tiny grappling hook from his pocket. "I brought the hook." Bill tugged a tiny smoke bomb. "I also brought confetti." Molly frowned. "That's not smoke." "It is festive smoke." "Are we stealing presents or launching a parade?" "Both, if the clown cooperates." Molly sighed and wrote a note. Bill pocketed his candy. Molly checked her clipboard. "Budget: two gummy bears." "We'll split them fairly." "Only if one is red." "Excellent. We ascend the jungle gym, rappel over the castle, and scoop every ribbon-tied box into the duffel." "What about parents?" "Blame the piƱata." Molly tossed a candy-colored decoy onto the patio table. "Misdirection." "That looks like a donkey having a seizure." "Perfect. Let's move before it becomes suspicious." Molly crept toward the swing set. Bill fumbled with a net bag. "Smoke pellets?" "We're stealing toys, Bill, not overthrowing France." "The plan calls for theatrical exit." "The plan calls for avoiding Aunt Linda." Molly nodded solemnly. "Aunt Linda is the true villain." They reached the fence. Children shrieked while a clown juggled rubber chickens. A toddler spotted them and screamed. "That isn't part of the heist," Bill whispered. "It's a hostage situation." "Is the hostage a cake?" "The cake is also a hostage." Molly grabbed a box with a unicorn. Bill swept three gifts into the bag. "Remember the rule." Molly adjusted her mask. "What rule?" Bill asked. "We take only unopened gifts." "That's generous." "It's ethical looting." "The presents have feelings." "Yours does." Molly grabbed the robot quickly. "It winked at me." "I'm a professional thief." "You're a professional child." "Take the big blue one." "I can't. It has expectations." Molly grabbed it anyway and stuffed the duffel very tight, then grinned with pride. The clown turned. Bill yanked the duffel over his shoulder. "Abort?" "Never." Molly snatched a present shaped like a shark. "Those are cookies." "All loot is loot." Bill dragged her away as the toddler launched a streamer like a missile. It wrapped around his face. He crashed into the bouncy castle and bounced upward, holding the gift bag like a parachute. Molly ran beneath him, catching falling presents. "This is a terrible plan," Bill shouted from above. "It's beautiful. You're elevating the operation," Molly called. Aunt Linda appeared with a spatula. Molly bowed. "Thank you for the party." Bill bounced past, yelling apologies. Molly tossed the shark cookie into a bush. Behind the recycling bin, they counted the loot. "Twenty-six gifts," Molly said. "Twenty-seven," Bill corrected. "The cursed yo-yo." "It followed me home." Molly examined a robot. "We're rich." Bill held up a stuffed cat. "This one winks." "Of course. That's our cut."
created_at
"2026-09-02T17:01:09.267Z"
9/2/2026, 5:01:09 PM
model
"qwen/qwen3.8-flash-20260826"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
695
moderation_latency
(null)
generation_time
189848
tokens_prompt
75
tokens_completion
16258
native_tokens_prompt
127
native_tokens_completion
22216
native_tokens_completion_images
(null)
native_tokens_reasoning
21411
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
0.01046057
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788368469-z0Q6kV4V95HZTqdlTUeI"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788368469-4dI9RYlCWHP7SjdHDUEV"
upstream_id
"chatcmpl-f466ae0a-8a95-9934-921d-0ce528a0f941"
provider_responses
0
endpoint_id
"84b1e4a7-5aed-4464-818f-1994f0b4ee18"
id
"chatcmpl-f466ae0a-8a95-9934-921d-0ce528a0f941"
is_byok
false
latency
695
model_permaslug
"qwen/qwen3.8-flash-20260826"
provider_name
"Alibaba"
status
200
total_cost
0.01046057
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
98.79%
Dialogue to Total Word Ratio
Ratio: 50.49%, Deviation: 0.49%
neededClean
false
wordsTotal
507
wordsDialogue
256
99.3958%