NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHYAVE8JQZXHGTKB34BRMC9P
Run Details
Model:
Gemma 3 12B
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
6.8s
Cost:
$0.000041
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
0.03%
Matches word count
(details)
100.00%
Dialogue to Total Word Ratio
(details)
50.0146%
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
“Right, so we’re clear. Clown costumes. Full clown regalia. Red nose, oversized shoes, the whole shebang,” Molly insisted, fiddling with a feather boa. Bill groaned. “Molly, I look dreadful in red. Absolutely dreadful. Can't we just go as… uh… disgruntled gardeners?” “Bill! Children’s party! Presents! Think of the haul! Diamond-encrusted tiaras, remote control cars, possibly a pony! Gardeners won’t get past the bouncy castle.” “A pony? Are you sure there’ll be a pony?” Bill squinted. “Highly likely. It’s Penelope Worthington’s birthday. Her father owns half the city. If there isn’t a pony, I’m blaming you.” “Fine, fine. Clown. But I'm drawing the nose on small. Very small.” “No arguing! And remember, distraction is key. You juggle. Badly. I’ll flutter around, complimenting the cake.” “Juggling? I haven't juggled since… well, never.” “Exactly! It'll look suspicious! They’ll be focusing on your appalling juggling skills, and we’ll be grabbing the loot!” Molly grinned, a glint in her eye. “Now, let’s find a wig that doesn’t make me look like a startled poodle!"
created_at
"2026-02-20T20:11:25.851Z"
2/20/2026, 8:11:25 PM
model
"google/gemma-3-12b-it"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
528
moderation_latency
(null)
generation_time
6692
tokens_prompt
93
tokens_completion
309
native_tokens_prompt
75
native_tokens_completion
293
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
0.00004109
router
(null)
provider_responses
0
endpoint_id
"eb06dc92-5a16-47ec-a776-6ef956457c47"
id
"chatcmpl-RAVeQ9RrEscdhN4grfRpMLiI"
is_byok
false
latency
100
model_permaslug
"google/gemma-3-12b-it"
provider_name
"DeepInfra"
status
200
api_type
"completions"
id
"gen-1771618285-ETDbi5uqkSYWcuIXUUwK"
upstream_id
"chatcmpl-RAVeQ9RrEscdhN4grfRpMLiI"
total_cost
0.00004109
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepInfra"
Evaluation details
Result
Evaluator
Details
Meta Data
0.03%
Matches word count
n/a
neededClean
false
words
170
100.00%
Dialogue to Total Word Ratio
Ratio: 89.41%, Deviation: 0.59%
neededClean
false
wordsTotal
170
wordsDialogue
152
50.0146%