NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1M004PFD6SSFYX1JAE3PRWC
Run Details
Model:
Qwen 3.8 27B (Reasoning, XHigh)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
4m 42s
Cost:
$0.033185
Provider:
AkashML
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 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 tapped the blueprints as Bill balanced a birthday candle on his nose. “First,” Molly said, “we enter through the clown car.” Bill slid it off his face. “The clown car is a metaphor.” “It is also a tent with a zipper.” Molly traced a crayon line across the lawn. “We go in as bouncy castle maintenance.” Bill nodded. “I am already wearing the official onesie.” “Excellent. I will play the technician.” “I will play the intern who smells of grape soda.” They crouched over a map drawn on a napkin. The party was at the Henderson house, behind a picket fence and a suspicious inflatable dinosaur. “The Hendersons expect twenty children,” Molly said. “Each child brings one present. Twenty presents is a lot of cardboard.” “Also ribbon,” Bill added. “The ribbon is structural.” Molly pointed to the napkin. “We must steal all the presents before the cake is cut.” Bill frowned. “What if the children ask why their gifts are gone?” “We tell them Santa had a delivery error.” “Santa does not deliver on Saturdays.” “Then it was a very early Santa.” Bill rolled a marblescope in his pocket. “My plan is simpler. We knock, compliment the frosting, and take the boxes.” “The frosting is a trap; traps are for bears, but we are thieves, not bears, so we must be cleverer than icing.” “How?” “It has sprinkles shaped like eyes.” Molly lifted the blueprints. “The parents will be in the kitchen. The children will be in the backyard. The presents are stacked beside the punch bowl, which is shaped like a volcano.” Bill swallowed. “A punch volcano.” “Do not let the tongs see you.” They heard a trombone from the neighboring yard. Bill straightened his onesie. “Rehearse the signal,” Molly said. He cleared his throat. “If I wiggle my left foot, you grab the Lego set.” “If you wiggle your right foot, I grab the stuffed dragon.” “If I wiggle both?” “You run.” Bill wiggled both. Molly sighed. “Okay. New signal. If I say ‘happy birthday,’ you open the trunk.” “The trunk is full of foil balloons.” “Still open it.” Molly drew another line. “We need a distraction. Something bright, loud, and terrible.” Bill pulled a kazoo from his collar. “I have prepared.” Molly stared. “That is not preparation. That is a cry for help.” “It has a little bow.” “Then we use the bow.” Bill lifted the kazoo with respect. “What does the bow do?” “It says we meant to look ridiculous.” Molly smoothed the napkin. “Listen. At two, you distract the clown. At three, I distract the cake. At four, we walk into the backyard.” Bill pointed to the inflatable dinosaur. “What about the dinosaur?” Molly said, “The dinosaur is our backup.” “The dinosaur is plastic.” “Exactly. It can be lifted, carried, and blamed for anything.” Bill clapped. “I like a plan with teeth.” “It does not have teeth.” “It has potential.” Molly folded the napkin and sealed it with a sticker of a smiling starfish.
created_at
"2026-09-03T16:00:30.422Z"
9/3/2026, 4:00:30 PM
model
"qwen/qwen3.8-27b-20260814"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
652
moderation_latency
(null)
generation_time
281951
tokens_prompt
75
tokens_completion
9854
native_tokens_prompt
118
native_tokens_completion
13259
native_tokens_completion_images
(null)
native_tokens_reasoning
9051
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
0.03318526
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788451230-F6MSpP9nUfwiO9l4oFjv"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788451230-TSAZXk6ZKifY1MC5fgj8"
upstream_id
"chatcmpl-8e6267176f881299"
provider_responses
0
endpoint_id
"c15e008c-6180-4155-8707-bde541bc7884"
id
"chatcmpl-8e6267176f881299"
is_byok
false
latency
652
model_permaslug
"qwen/qwen3.8-27b-20260814"
provider_name
"AkashML"
status
200
total_cost
0.03318526
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
100.00%
Dialogue to Total Word Ratio
Ratio: 70.00%, Deviation: 0.00%
neededClean
false
wordsTotal
500
wordsDialogue
350
100.0000%