NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHY8RQBSKARVGQ4J0BWRR0F5
Run Details
Model:
Gemma 3 4B
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
4.5s
Cost:
$0.000023
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0000%
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
“Right, so, Operation Sugar Rush. We’re in, we grab the loot, we’re out. Simple,” Molly said, meticulously arranging miniature plastic dinosaurs on the kitchen table. Bill chewed on a pen cap. “Simple for *you*, maybe. I’m picturing a room full of screaming toddlers and a sugar-fueled security guard named Barry. Barry’s a problem.” “Barry’s predictable. He’ll be glued to the bouncy castle, hypnotized by the screaming. Just move through the gift table. Minimal contact, maximum presents.” “Minimal contact? Molly, these aren't diamonds. They’re Paw Patrol figurines and sparkly unicorn headbands.” “Exactly! Think of the resale value! Plus, little Timmy’s got that limited edition Justin Bieber plush, that’s worth a fortune.” “You’re seriously prioritizing Bieber over, you know, not getting arrested?” “Look, Barry’s distracted by a piñata. I’ve mapped the escape route – through the crepe paper and out the emergency exit. Just grab and go. And Bill?” “Yeah?” “Don't touch the frosting."
created_at
"2026-02-20T19:34:59.715Z"
2/20/2026, 7:34:59 PM
model
"google/gemma-3-4b-it"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
199
moderation_latency
(null)
generation_time
4401
tokens_prompt
93
tokens_completion
287
native_tokens_prompt
75
native_tokens_completion
250
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
0.000023
router
(null)
provider_responses
0
endpoint_id
"d3de3bd4-81bc-48fb-924f-2a87b2a36e75"
id
"chatcmpl-R2hiF70qMzKxnLfHViA04qOn"
is_byok
false
latency
65
model_permaslug
"google/gemma-3-4b-it"
provider_name
"DeepInfra"
status
200
api_type
"completions"
id
"gen-1771616099-uGI8ukAUjtBYjVz4ONCc"
upstream_id
"chatcmpl-R2hiF70qMzKxnLfHViA04qOn"
total_cost
0.000023
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
152
0.00%
Dialogue to Total Word Ratio
Ratio: 88.82%, Deviation: 38.82%
neededClean
false
wordsTotal
152
wordsDialogue
135
0.0000%