NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1M1630BEC8GT9V8032GRY1Q
Run Details
Model:
Qwen 3.8 27B (Reasoning, XHigh)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
1m 23s
Cost:
$0.030118
Provider:
Venice
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
“We enter through the toy chute and collect all presents.” Molly tucked a crayon lockpick into her sleeve and grinned. “Bill, we ride the bounce castle to the gift table.” Bill strapped two candy grenades over his tactical tutu straps. “Every doll becomes contraband, every robot becomes payroll, every cake.” The party banners showed cupcakes, but the vault hid drones. “Check the juice bar, yes, it masks our scanners, Bill.” She polished her balloon grappling hook beneath a glitter veil. “Our inside contact wears a paper crown and leaks schedules.” Bill counted balloons while Molly traced sprinkler wiring through napkins. “Use napkin sleeves for fingerprints, then pop streamers for smoke.” He nodded, pretending their van looked merely like a cake. “The cake van opens with icing, never keys, genius, Bill.” A clown alarm would scream if any doll crossed velvet. “We disable it with lullabies, then bag every present quickly.” Children would applaud, mistaking thieves for birthday performers in wigs. “Molly, your wig has glitter, will cameras read our sparkle?” Bill tapped Molly, her fake nose chirped like a cricket. “Ignore cameras, they only love glitter, not theft, Bill, please.” The vault door had frosting hinges, birthday confetti locks, squeaks. “My lockpick tastes licorice, your scanner smells lemon, perfect, partner.” Outside, police drones wore hats, humming badly near the slides. “They will arrest whoever holds the biggest gift, us, Molly.” Bill loaded toy bricks into his belt like tiny bricks. “Bricks pay better than jewels, because they multiply after midnight.” Midnight would mean cake, cake meant chaos, chaos meant profit. “Split presents evenly, but Molly takes the golden unicorn, yes.” Molly raised her wand, Bill raised two foam pistols shaped. “Shaped like baguettes, because Paris called, and Paris needs bread.” A toddler watched, holding a balloon that mapped security cameras. “That balloon is our map, Bill, follow its knot, now.” He followed, stepping over plastic swords, dodging spilled punch, hats. “Don't let your hat touch sprinklers, or static will betray.” Static would smell of lemon, lemon would attract security puppies. “Puppies? We have puppies, Bill, they guard stuffed bears here.” The bears had microphones, Bill sighed, Molly smiled, ready now. “Ready now, let's dance, let's run, let's pocket toys loudly.” Loudly they slid down a chute full of plush sentries. “Sentries snore, good, we need silence, Bill, hush now, please.” Silence became tape, security tape across the gift tables, yes. “Cut the tape slow, else the tape sings to guards.” Bill hummed while Molly cut, sparks flew like tiny stars. “Stars for presents, yes, we own stars, Bill, dear Molly.” The unicorn sat atop the gifts, glowing softly under icing. “Grab the unicorn, Bill, it pays rent, it pays lawyers.” Bill grabbed, Molly bagged, robots squeaked, dolls clapped, cake grinned. “Cake grins, alarm laughs, we must hurry, Molly, now yes.” The alarm laughed, clown horns squealed, hats spun into smoke. “Through the cake, yes, Bill, escape, presents, laughter, us now.” They emerged, laughing, stuffed shirts, glitter, gifts, stars, now free.
created_at
"2026-09-03T16:21:13.874Z"
9/3/2026, 4:21:13 PM
model
"qwen/qwen3.8-27b-20260814"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
528
moderation_latency
(null)
generation_time
83301
tokens_prompt
75
tokens_completion
7492
native_tokens_prompt
155
native_tokens_completion
9390
native_tokens_completion_images
(null)
native_tokens_reasoning
6647
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
0.03011775
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788452473-T7nOjtu0Fab10nFJ6pFl"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788452473-b49v3thqJE9wqmWKWLaG"
upstream_id
"chatcmpl-ada27e9f4d57345a"
provider_responses
0
endpoint_id
"96058344-d6fa-4cfe-a99f-dbce2b53c90a"
id
"chatcmpl-ada27e9f4d57345a"
is_byok
false
latency
528
model_permaslug
"qwen/qwen3.8-27b-20260814"
provider_name
"Venice"
status
200
total_cost
0.03011775
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Venice"
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%