NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1M2HHMJAPSTX54ZGX4HYC0Z
Run Details
Model:
Qwen 3.8 27B (Reasoning, XHigh)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
4m 58s
Cost:
$0.052095
Provider:
AkashML
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
99.72%
Dialogue to Total Word Ratio
(details)
99.8585%
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 crouched behind a topiary. "Bill, if the birthday boy sneezes, do you panic or deploy?" Bill peeked over the hedge. "Panic, because my laser cutter is in the other van." "You packed the laser cutter in the ice cream van?" "It was a tactical decision." "Mostly, you wanted the sprinkles." "Sprinkles are morale." Molly tapped the clipboard. "Presents count. Thirty-seven boxes. Forty-two tags. One suspiciously large dragon." "The dragon is bait." "The dragon is the goal. Look, it has wheels." "Then we need a heist with wheels." "The dragon has a name tag." "Can we steal the dragon without a plan?" "We have a plan." "Which plan?" "The one where we dress as party favors." "We are not wearing glitter." "It will make us invisible." "It will make us sticky." "The parents will think we are expensive." Molly pointed to the front door. "The mother is counting candles." "Let her count. Candles are not presents." "Those candles are on the cake. The cake is on a table with presents." "Good. We take the table." "Can you lift a table with thirty-seven gifts?" "Can you open a box with forty-two tags?" "I can open one with a hairpin and dignity." "Dignity does not unzip a dinosaur." Bill removed a wrench from his sleeve. "I bring tools." "You brought a wrench to a children’s party." "I respect the venue." "Your wrench is still in your pocket." "Then trust the process." Molly opened the gate. "Process, please." They crept along the balloon border. "If we steal all the presents, will the kids scream?" "They will cry." "Then we leave one present." "Which one?" "The one with the biggest ribbon." "The dragon?" "Exactly." "No, the dragon is ours." Molly stopped beside a potted flower. "Bill, the birthday boy is inside." "He sees us." "We are party favors." "Party favors are not six feet tall." "We are a premium set." The front door opened. A child in dinosaur pajamas appeared holding a lollipop. Molly smiled. "We are very rare and extremely hard to return." Bill nodded. "Limited edition, no warranty card." The child pointed to the table. Molly patted Bill. "He is already in on our silly party heist." Bill whispered through a grin. "We have to be fast, sneaky, and very polite." The child laughed and rolled presents toward them. Molly stared at the table. "He is giving us the presents?" Bill looked at the table. "Maybe the children are smarter than we are." "Maybe the children are lazier." "Maybe the heist is easy." Molly grabbed the dragon. "Then let us take the easy." Bill grabbed a box. "The easy has wheels." The party music started. "Wait, is that the alarm?" "No," said Molly, rolling the dragon toward the van. "That is the party." Bill locked the van, checked the tire pressure, polished the dragon’s shiny wheels, and made sure every single present was strapped tightly inside before he glanced back at the silent party house and winked at the sleepy moon. "Done."
created_at
"2026-09-03T16:44:57.939Z"
9/3/2026, 4:44:57 PM
model
"qwen/qwen3.8-27b-20260814"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
332
moderation_latency
(null)
generation_time
297860
tokens_prompt
75
tokens_completion
18348
native_tokens_prompt
118
native_tokens_completion
23666
native_tokens_completion_images
(null)
native_tokens_reasoning
17605
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
0.0520947
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788453897-kdo2Qx8ZOjLFuZ6O7N2Y"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788453897-3sDR9q6QPeFVjlELdVdk"
upstream_id
"chatcmpl-975ba7161b5b3606"
provider_responses
0
endpoint_id
"c15e008c-6180-4155-8707-bde541bc7884"
id
"chatcmpl-975ba7161b5b3606"
is_byok
false
latency
327
model_permaslug
"qwen/qwen3.8-27b-20260814"
provider_name
"AkashML"
status
200
total_cost
0.0520947
cache_discount
(null)
upstream_inference_cost
0
provider_name
"AkashML"
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
99.72%
Dialogue to Total Word Ratio
Ratio: 70.24%, Deviation: 0.24%
neededClean
false
wordsTotal
504
wordsDialogue
354
99.8585%