NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-25-01KMJ8KJEMGWE8SHE6M50223JH
Run Details
Model:
Qwen 3 32B
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
33.6s
Cost:
$0.000708
Provider:
SiliconFlow
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
7.00%
Dialogue to Total Word Ratio
(details)
3.5001%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 90% 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 a potted fern, her trench coat stained with suspiciously glittery mud. "Bill, we’re talking *Operation Candy Caper*. The parents will be too busy sipping mimosas to notice a thing." Bill adjusted his fake mustache, which had already slipped over his nose. "But it’s a *kids’* party. There’s a bounce house, a magi—wait, *why* is there a magician? Did we sign up for this?" "Relax," Molly said, tossing him a neon hat. "We’ll discretely—*ahem*—’redeploy’ the gifts while he’s doing his ‘poof’ thing. Now, disguises. You’re ‘Dance Captain Derek.’ I’m ‘Elf Extraordinaire’… which is code for I’m wearing Christmas pajamas and you’re not allowed to laugh." Bill glared, holding up a balloon animal shaped like a cobra. "This is a *clown*, Molly. A clown! What if they have *two* clowns? Or a *disguised* clown? Or a clown *dressed* as a magician?" Molly handed him a whoopee cushion. "First rule of stealing from tots: never question the chaos. Second rule: don’t get glitter in your socks. Third—” "Wait, the presents! Are we tackling the *stocking* stash or the *under-the-tree* vault? Because if it’s the latter, we’re gonna have to negotiate with Santa’s little… *helpers*." "Glitter bombs," Molly said bluntly. "We’ll deploy them near the cake. Kids freak out, parents trip, we slip in, swap the Barbies for fake ones, and disappear before someone realizes their kid’s drawing on a plexiglass wall with a glow stick." Bill whimpered. "What if we get tackled by a toddler? I can’t explain fingerprints in a Tupperware container!" "Simple." Molly grinned, revealing a sparkly fanny pack. "We’ll bribe the evidence with bubble gum. Now, synchronize watches… and try not to cry when the piñata refuses to break open." Bill stared at her. "You knew there was a piñata?" "It’s a *sign*. The universe loves a good heist. Just… try not to let the ‘baby’ in the onesie cling to you for dear life, okay?" ".setOnion breath? What if it’s a *fire* piñata?" Molly clapped. "You’re learning!" Bill sighed. "I miss robbing banks." "Next job," Molly said, tossing him a noisemaker, "we hit a *toddler* *bank*. Cash, candy, and zero accountability." "Glitter," Bill muttered, "is a *weapon of mass destruction*." "Roll out, Derek." Molly tossed him the balloon cobra. "And *smile*—you’re a thief with a sentimental side." Bill lobbed the cobra back. "I hate this plan." "Love it," Molly said, already sprinting toward the bounce house. "And hurry up before someone finds my ‘elf’ costume’s hidden pocket! *This* isn’t where the toys go, Bill. *This* is where the *real* loot goes."
created_at
"2026-03-25T10:28:23.128Z"
3/25/2026, 10:28:23 AM
model
"qwen/qwen3-32b-04-28"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1277
moderation_latency
(null)
generation_time
33493
tokens_prompt
93
tokens_completion
1319
native_tokens_prompt
72
native_tokens_completion
1225
native_tokens_completion_images
(null)
native_tokens_reasoning
510
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
0.00070833
router
(null)
provider_responses
0
endpoint_id
"6f9e09e5-e836-4de0-a27d-27c74b866001"
id
"019d2489cd6c33906511a78173ba3477"
is_byok
false
latency
1260
model_permaslug
"qwen/qwen3-32b-04-28"
provider_name
"SiliconFlow"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.0; linux; x64))"
http_referer
(null)
api_type
"completions"
id
"gen-1774434503-e7gaGUhoU8nzUKrdrww8"
upstream_id
"019d2489cd6c33906511a78173ba3477"
total_cost
0.00070833
cache_discount
(null)
upstream_inference_cost
0
provider_name
"SiliconFlow"
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
426
7.00%
Dialogue to Total Word Ratio
Ratio: 77.23%, Deviation: 12.77%
neededClean
false
wordsTotal
426
wordsDialogue
329
3.5001%