NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1M11V9J2N223ZDVX6PAYWJ5
Run Details
Model:
Qwen 3.8 27B (Reasoning, XHigh)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
7m 41s
Cost:
$0.084480
Provider:
Novita
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 spread the party map across the picnic table, pretending she was not excited. "Bill, this party is a present factory," Molly said. "Every gift is wrapped in plastic, every child is sticky, and the host forgot to lock the side door." Bill adjusted his hat. "Side door," Bill said. "I like side doors. They make me feel small and dangerous." "We need a distraction," Molly continued. "A clown. A tiny clown with a giant nose and a weak sense of direction." Bill laughed. "A weak sense of direction?" Bill asked. "The last clown you hired followed us into a parking garage." "That clown was a single mistake," Molly said. "Also, he had bad eyebrows." "Eyebrows can be useful," Bill said. "They can hide a stolen hat." Molly pointed at the map. "The presents are here, by the bouncy castle," she said. "The cake is there. The parents are over there, taking pictures of balloons." "Pictures of balloons," Bill repeated. "Perfect. Balloons make parents forget everything." "They do not forget everything," Molly said. "They just forget what they were doing." Bill picked up a donut bag. "This is our bag," he said. "It can hold one hundred gifts or forty juice boxes, whichever appears first." "Forty juice boxes?" Molly asked. "Where are we putting forty juice boxes?" "The van," Bill said. "It is full of suspiciously empty space." Molly checked her watch. "We enter at two," she said. "The clown arrives at two fifteen. We move during the first song, before the kids learn the words." "First song?" Bill asked. "Which first song?" "The one with the foam sword," Molly said. Bill shook his head. "You cannot plan a heist around foam," he said. "Foam is soft. Heists need hard things." "Soft things distract children," Molly said. "Hard things frighten parents." "And presents?" Bill asked. "Presents are both," Molly said. Molly held up a tiny plastic knife. "For the cake," she said. "We do not need to steal the cake. We only need to pretend we might." "Pretend?" Bill asked. "I am very good at pretending. My mother says I look like a birthday guest." Molly tapped the map again. "When we have the gifts," she said, "we exit through the fountain." "The fountain?" Bill said. "The fountain has turtles." "Turtles are slow," Molly said. "Those turtles have a record for fast swimming," Bill said. "Then we go around the fountain," Molly said. "Around the fountain," Bill repeated. "That is just another fountain." "It is decorative," Molly said. "Decorative is my enemy," Bill said. Molly placed the knife in Bill's pocket and lifted the donut bag onto her shoulder. She checked the party map, then tucked it inside her blazer. The sun shone on the inflatable castle. Two balloons drifted slowly near the window outside. "Last chance to panic, partner. We steal every gift, then vanish before the piƱata explodes." "Perfect. I have a plan for the fountain, the turtles, and the decorative window." "Finally, we are both ready, partners."
created_at
"2026-09-03T16:18:54.907Z"
9/3/2026, 4:18:54 PM
model
"qwen/qwen3.8-27b-20260814"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
545
moderation_latency
(null)
generation_time
460333
tokens_prompt
75
tokens_completion
22204
native_tokens_prompt
114
native_tokens_completion
28144
native_tokens_completion_images
(null)
native_tokens_reasoning
27415
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
0.08447988
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788452334-7vC13ouDE1iNArEeE7Ha"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788452334-zVGcUqjnJKijxCROzK1t"
upstream_id
"7e90a47119734ef4892179b8febeb002"
provider_responses
0
endpoint_id
"2af35fbf-0463-4d96-96d1-1e82b1a46dc7"
id
"7e90a47119734ef4892179b8febeb002"
is_byok
false
latency
545
model_permaslug
"qwen/qwen3.8-27b-20260814"
provider_name
"Novita"
status
200
total_cost
0.08447988
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Novita"
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%