NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-11-01KKFFT0GVRH8DEA0RYQ0330NG
Run Details
Model:
Qwen 3.5 9B
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
3m 59s
Cost:
$0.002364
Provider:
Together
Evaluation summary
Score
Evaluator
Details Link
99.38%
Matches word count
(details)
99.72%
Dialogue to Total Word Ratio
(details)
99.5469%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 10% 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
Moonlight slowly bled through the hydrangeas surrounding the sparkling suburban villa. Molly adjusted her mask, ensuring the silk she did not actually need didn't catch a single wisp of dust. Bill stood by the window, squinting at the open toy display organized inside the room. Toy blocks stacked like Tetris ruins. Plush dragons slept. The target was the presents under the tree, wrapped in colorful paper like sweet candy. They needed precision. Bill stood silent. Bill whispered, "He's leaving the back door standing open." Molly nodded lightly. She meant the man inside the house. This wasn't a heist for bank codes. This was about tampering with twisted geometry. Socks were matching, not plastic mesh. The dad. Bill moved his hand over the mask. "Do you actually have any siblings? come quickly." Bill moved. He chose to steal now. No joking. They slipped into the hallway. The party continued. No one had noticed them yet. But the cake smelled better than jewels. Molly shrugged. Then Bill grabbed a box of balloons. It was fluff. The loot was soft. The dad appeared. He breathed softly. It was foggy here, but truly not really. Yes. Okay. Go. Stay.
created_at
"2026-03-11T22:20:49.065Z"
3/11/2026, 10:20:49 PM
model
"qwen/qwen3.5-9b-20260310"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
5699
moderation_latency
(null)
generation_time
238371
tokens_prompt
77
tokens_completion
11775
native_tokens_prompt
76
native_tokens_completion
15708
native_tokens_completion_images
(null)
native_tokens_reasoning
11485
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
0.0023638
router
(null)
provider_responses
0
endpoint_id
"00d506d9-a5e9-4382-a2b9-d54732365594"
id
"oagbDRy-3pDw3Z-9dadf2d6c0c44766"
is_byok
false
latency
5699
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-1773267649-QWPTZBTiE5LAj2IVwb2J"
upstream_id
"oagbDRy-3pDw3Z-9dadf2d6c0c44766"
total_cost
0.0023638
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Together"
Evaluation details
Result
Evaluator
Details
Meta Data
99.38%
Matches word count
n/a
neededClean
false
words
195
99.72%
Dialogue to Total Word Ratio
Ratio: 7.69%, Deviation: 2.31%
neededClean
false
wordsTotal
195
wordsDialogue
15
99.5469%