NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-25-01KMJ8W51JWY5Z9AM9663AFXTD
Run Details
Model:
Qwen 3 32B
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
13.9s
Cost:
$0.000251
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
99.64%
Dialogue to Total Word Ratio
(details)
49.8175%
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 adjusted the sequined cuffs of her clown costume while eyeing the inflatable tiger bouncing in the backyard. Bill, wearing a mustache that drooped mournfully, held a grocery bag labeled *"Party Favors: All the Glitter, None of the Joy."* “This is insane,” Bill said, ducking as a child’s piñata swung overhead. “We’re stealing toys from *toddlers*.” “We’re upgrading their toys,” Molly corrected, slapping a whoopee cushion on a lawn chair. “These ‘toys’ are creaky, mass-produced junk. We’re giving them *character*. A teddy bear that yells dirty riddles? That’s a legacy.” Bill tugged at his mustache. “And if the parents notice?” “They’re too busy eating the cake shaped like a talking alphabet cookie. Now, follow the balloon train. When the face painter distracts the kids, we swap the presents.” Bill peered at a tower of wrapped boxes. “What if someone *hugs* us? I can’t fake a high-five and palm a gift card.” Molly tossed him a neon tambourine. “Then don’t. Just keep jingling. Also, if you trip over the moonwalk, I’m leaving you there.” A toddler waddled toward Bill, clutching a balloon. “Bebo, want sticker?” Bill stared, frozen, as Molly whisper-shouted, “That’s a *code name*, move!” As Bill backed away, he knocked over a punch bowl. The resulting splash drew a cheer from the children. Molly facepalmed. “Remember,” she hissed, “theft is a service. Now make like a piñata and *don’t break.”* They slipped toward the presents, Molly’s clown shoes squeaking triumphantly.
created_at
"2026-03-25T10:33:04.311Z"
3/25/2026, 10:33:04 AM
model
"qwen/qwen3-32b-04-28"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
106
moderation_latency
(null)
generation_time
13825
tokens_prompt
93
tokens_completion
990
native_tokens_prompt
73
native_tokens_completion
875
native_tokens_completion_images
(null)
native_tokens_reasoning
562
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
0.00025084
router
(null)
provider_responses
0
endpoint_id
"6b8c829d-3094-45e7-8139-0a67e09060c3"
id
"cmpl-9e3c189454e942309e1947e55faf6182"
is_byok
false
latency
46
model_permaslug
"qwen/qwen3-32b-04-28"
provider_name
"DeepInfra"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.0; linux; x64))"
http_referer
(null)
api_type
"completions"
id
"gen-1774434784-ciisKNFszeizOUmqSHkd"
upstream_id
"cmpl-9e3c189454e942309e1947e55faf6182"
total_cost
0.00025084
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepInfra"
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
241
99.64%
Dialogue to Total Word Ratio
Ratio: 52.46%, Deviation: 2.46%
neededClean
false
wordsTotal
244
wordsDialogue
128
49.8175%