NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-02-01M1HH4XHQKZG8RZ5PXK1A2D1Z
Run Details
Model:
Qwen 3.8 Flash (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
2m 31s
Cost:
$0.007383
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 crouched behind a hedge shaped like a confused unicorn. She checked the time and studied the birthday party spread across the lawn. Folding tables groaned under wrapped presents. Balloons bounced against fences. A clown dozed near a cooler. The afternoon sunshine made everything seem innocent, including the crocodile backpack. "The plan is simple. We enter the party quietly. We steal the presents quickly. We avoid the bouncy castle. We never fight a toddler. We distract the parents with cake. We leave before the clown wakes up. Do you understand? Put on the gloves. Look serious. Smile. We are pros." Molly rolled her eyes and adjusted a fake mustache. Bill tugged at his black gloves and stared at the children as if they owed him money. He had rehearsed this plan for three days, which mostly meant rehearsing complaints. The target table remained bright, tempting, and unguarded all day again. "No. I do not like this. We rob adults. We rob banks. We do not rob children. That table has a dinosaur cake and seventeen gifts. If caught, I will be laughed out of every villain guild. I may never be invited to a rooftop heist. I will not reconsider." "You are being dramatic. Children get presents constantly. They will not miss a few boxes. We will be back before the cake. Do you remember the training? We have a diversion. You have the crocodile backpack. I have the lollipop bribe. Everything is arranged. Everything is inevitable. Everything is ours." Bill looked as if the word inevitable had personally insulted his family. A balloon drifted between them and burst against the fence, making him flinch. Molly did not flinch. She had spent years flinching in dignified places, including rear windows of vans. The backyard remained oblivious, cheerful, and still exposed. "You forgot the baby monitor. You forgot the grandmother. You forgot the dog with the suspicious collar. What if a child screams? What if the cake collapses? What if my mustache falls off? I am not ready for this level of emotional damage. I also want therapy, insurance, and snacks." Molly ignored the snacks comment because she knew Bill would never share pudding. She pointed to the fern behind the cake table, their escape route. The plan depended on timing, misdirection, and the law that no one watches a fern while someone nearby holds a cupcake. It was stupidly brilliant. "Then stop talking and start believing. The presents are already ours. We simply have to collect them like reluctant librarians. On my signal, you lift the biggest box. I will release the candy cane distraction. Bill, look at me. You are dangerous. You are mysterious. You are mildly unattractive now." Bill straightened, offended but encouraged. He took the crocodile backpack and checked the candy canes as if they were sugary explosives. The party music crackled through a speaker. A child shouted for slime. Molly raised three fingers, counting down to the most childish robbery in history, and they were ready.
created_at
"2026-09-02T17:02:44.655Z"
9/2/2026, 5:02:44 PM
model
"qwen/qwen3.8-flash-20260826"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
775
moderation_latency
(null)
generation_time
136209
tokens_prompt
75
tokens_completion
13403
native_tokens_prompt
127
native_tokens_completion
15667
native_tokens_completion_images
(null)
native_tokens_reasoning
15024
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
0.00738254
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788368564-9Z33I2865Vr8UYcd3nJw"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788368564-EbXwgxzrRHrAoAWMCVv8"
upstream_id
"chatcmpl-0e78b512-3d32-9305-81b2-9373fd48f734"
provider_responses
0
endpoint_id
"84b1e4a7-5aed-4464-818f-1994f0b4ee18"
id
"chatcmpl-0e78b512-3d32-9305-81b2-9373fd48f734"
is_byok
false
latency
775
model_permaslug
"qwen/qwen3.8-flash-20260826"
provider_name
"Alibaba"
status
200
total_cost
0.00738254
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%