NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-25-01KMJ92Z83M6Z3BSPV18ZDMNEC
Run Details
Model:
Qwen 3 32B
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
8.2s
Cost:
$0.000340
Provider:
AtlasCloud
Evaluation summary
Score
Evaluator
Details Link
68.10%
Matches word count
(details)
42.99%
Dialogue to Total Word Ratio
(details)
55.5457%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 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
Molly crouched behind the inflatable unicorn in the corner of the bounce house, squinting at a table stacked with brightly wrapped boxes. "Bill," she whispered, tugging on his sleeve, "are you *sure* this is a good idea? We’re about to rob a party for six-year-olds." Bill adjusted his green party hat — which was definitely not undercover gear — and grinned. “It’s not *robbing*, it’s… strategic redistribution. These boxes are worth thousands in high-end toymarkets.” "They're just toys," Molly muttered. "Plush unicorns made of memory foam,” Bill countered. “And that remote-controlled dinosaur over there? It has Wi-Fi connectivity and Alexa integration. That’s pure gold.” A cheer went up from a group of kids across the room as a clown started juggling confetti. Molly cringed. “If they so much as *suspect* we’re not here to play spin the bottle, we’re history.” “Relax,” Bill said, tossing her a pair of noise-cancelling earbuds. “I got us kid-formulas. We blend right in.” “If we ‘blend right in,’ why do I have glitter in my hair?” Bill winked. “Part of the disguise.” "You’re an idiot." “And yet, here you are, still playing along.”
created_at
"2026-03-25T10:36:47.751Z"
3/25/2026, 10:36:47 AM
model
"qwen/qwen3-32b-04-28"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1715
moderation_latency
(null)
generation_time
8199
tokens_prompt
93
tokens_completion
324
native_tokens_prompt
76
native_tokens_completion
277
native_tokens_completion_images
(null)
native_tokens_reasoning
0
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
0.00034
router
(null)
provider_responses
0
endpoint_id
"2693bf9a-1732-4008-acc9-d241214e936c"
id
"294c5086c52b450c9de63aac63b91dca"
is_byok
false
latency
1715
model_permaslug
"qwen/qwen3-32b-04-28"
provider_name
"AtlasCloud"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.0; linux; x64))"
http_referer
(null)
api_type
"completions"
id
"gen-1774435007-fGPS0Dla8YWvdvSInRRb"
upstream_id
"294c5086c52b450c9de63aac63b91dca"
total_cost
0.00034
cache_discount
(null)
upstream_inference_cost
0
provider_name
"AtlasCloud"
Evaluation details
Result
Evaluator
Details
Meta Data
68.10%
Matches word count
n/a
neededClean
false
words
186
42.99%
Dialogue to Total Word Ratio
Ratio: 59.59%, Deviation: 9.59%
neededClean
false
wordsTotal
193
wordsDialogue
115
55.5457%