NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-11-01KKFD6PY3MBQQWKBA42TQG5EN
Run Details
Model:
Qwen 3.5 9B
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
2m 9s
Cost:
$0.001391
Provider:
Together
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.94%
Dialogue to Total Word Ratio
(details)
0.4712%
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
The car engine hummed in the dark, a low growl like a sleeping beast hidden under the cold grass. Bill adjusted the goggles while Molly checked the box. "It looks safe," Bill said. "It looks locked," Molly replied. "Locked is good," Bill said. "Locked is hard," Molly argued. "Hard," Bill agreed. "Hard is safe." "Safe is good," Molly said. "Good," Bill said. "Good is big." "Big," Molly said. "Big is heavy." "Heavy," Bill replied. "Heavy is heavy." "Heavy," Molly said. "Heavy is bad." "Bad," Bill said. "Bad is no." "No," Molly said. "No is small." "Small," Bill said. "Small is light." "Light," Molly said. "Light is easy." "Easy," Bill said. "Easy is fast." "Fast," Molly said. "Fast is fun." "Fun," Bill said. "Fun is loud." "Loud," Molly said. "Loud is bad." "Bad," Bill said. "Bad is no." "No," Molly said. "We need toys." "Toys," Bill said. "Toys are big." "Big," Molly said. "Big is heavy." "Heavy," Bill said. "Heavy is hard." "Hard," Molly said. "Hard is not us." "Not us," Bill replied. "We are hard." "We are hard," Molly said. "Hard is tough." "Tough," Bill said. "Tough is good." "Good," Molly said. "Good is plan." "Plan," Bill said. "Plan is tools." "Tools," Molly said. "Tools are many." "Many," Bill said. "Many is too much." "Too much," Molly said. "Too much is bad." "Bad," Bill said. "Bad is bad." "Bad," Molly said. "Bad is plan." "Plan," Bill said. "Plan is silver." "Silver," Molly said. "Silver is light." "Light," Bill said. "Light is quiet." "Quiet," Molly said. "Quiet is good." "Good," Bill said. "Good is gold." "Gold," Molly said. "Gold is shine." "Shine," Bill said. "Shine is night." "Night," Molly said. "Night is dark." "Dark," Bill said. "Dark is safe." "Safe," Molly said. "Safe is good." "Good," Bill said. "Good is go." "Go," Molly said. "Go now." "Now," Bill said. "Now is here." "Here," Molly said. "Here is plan." "Plan," Bill said. "Plan is win." "Win," Molly said. "Win is ours." "Ours," Bill said. "Ours is home." "Home," Molly said. "Home is safe." "Safe," Bill said. "Safe is plan." "Plan," Molly said. "Plan is now." "Now," Bill said. "Run." "Run," Molly said. "Go." "Go," Bill said. "Run." The grass was wet under the tires. The porch light flickered. Bill checked the watch. "Watch is time," Bill said. "Time," Molly said. "Time is here." "Here," Bill said. "We go inside." "Inside," Molly said. "Inside is the gifts." "Gifts," Bill said. "Gifts are caps." "Caps," Molly said. "Caps are boxes." "Boxes," Bill said. "Boxes are full." "Full," Molly said. "Full is good." "Good," Bill said. "Good is plan." "Plan," Molly said. "Plan is win." "Win," Bill said. "We go home." "Home," Molly said. "Home is safe." "Safe," Bill said. "Safe is good." "Good," Molly said. "Good is gold." "Gold," Bill said. "Gold is shine." "Shine," Molly said. "Shine is night." "Night," Bill said. "Night is dark." "Dark," Molly said. "Dark is safe." "Safe," Bill said. "Safe is plan." "Plan," Molly said. "Plan is now." "Now," Bill said. "Ready." "Ready," Molly said. "We go." "Go," Bill said. "Go." "We go," Molly said. "We go now." "Now," Bill said. "Run." "Run," Molly said. "Go." "Go," Bill said. "Run."
created_at
"2026-03-11T21:35:19.503Z"
3/11/2026, 9:35:19 PM
model
"qwen/qwen3.5-9b-20260310"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
327
moderation_latency
(null)
generation_time
128372
tokens_prompt
77
tokens_completion
8568
native_tokens_prompt
76
native_tokens_completion
9225
native_tokens_completion_images
(null)
native_tokens_reasoning
7635
native_tokens_cached
0
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
usage
0.00139135
router
(null)
provider_responses
0
endpoint_id
"00d506d9-a5e9-4382-a2b9-d54732365594"
id
"oagMDy9-6Ng1vN-9dadb03300de45fe"
is_byok
false
latency
327
model_permaslug
"qwen/qwen3.5-9b-20260310"
provider_name
"Together"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.0; linux; x64))"
http_referer
(null)
api_type
"completions"
id
"gen-1773264919-emxGEiByJRfhlvTfMPmq"
upstream_id
"oagMDy9-6Ng1vN-9dadb03300de45fe"
total_cost
0.00139135
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Together"
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
532
0.94%
Dialogue to Total Word Ratio
Ratio: 60.34%, Deviation: 9.66%
neededClean
false
wordsTotal
532
wordsDialogue
321
0.4712%