NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHY8DARNQKAWS85K09HFHNTE
Run Details
Model:
Gemma 3 4B
Test:
Dialogue tags
Scenario:
Write unattributed dialogue
Duration:
2.9s
Cost:
$0.000024
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
13.53%
Count dialogue tags
(details)
Model input
User/Human
You are an expert novel writer in the cozy mystery genre. Write a scene of about 200 words, with dialogue between two characters. The dialogue should be unattributed, meaning the speaker is not explicitly mentioned. The scene should be set in a small town bakery. Only return the story, nothing else.
Model response
Messages
Meta
AI
The scent of cinnamon and warm bread hung thick in “The Sweet Spot,” clinging to everything like a comforting blanket. Flour dusted the counter, and a half-finished batch of apple tarts sat cooling on a rack. “You’re staring at the sugar sprinkles, Detective Harding,”
a voice observed
, laced with a touch of amusement. “Not exactly a crime scene investigation technique, is it?” “Just…observing,”
Harding replied
, pushing a stray lock of graying hair behind his ear. He took a slow sip of his coffee, the ceramic warm against his palm. “The way Mrs. Gable arranged them. It’s… deliberate.” “Deliberate? She sprinkles them randomly when she’s baking. It’s a Gable thing.” A pause, then, “Though, I *did* notice she used an unusual amount today. Almost like she wanted to draw attention to them.” “Exactly. And the delivery man, Mr. Henderson, he left without saying a word. Just dropped off a box of croissants and vanished.” “Henderson? He’s been a fixture here for twenty years. Something’s definitely off. Maybe we should ask about the sprinkles.” “Let’s start with the croissants.”
created_at
"2026-02-20T19:28:46.367Z"
2/20/2026, 7:28:46 PM
model
"google/gemma-3-4b-it"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
187
moderation_latency
(null)
generation_time
2718
tokens_prompt
98
tokens_completion
305
native_tokens_prompt
76
native_tokens_completion
261
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
0.00002392
router
(null)
provider_responses
0
endpoint_id
"d3de3bd4-81bc-48fb-924f-2a87b2a36e75"
id
"chatcmpl-Rl8nLdIQS6qnJ8Lb6iRpEQdr"
is_byok
false
latency
74
model_permaslug
"google/gemma-3-4b-it"
provider_name
"DeepInfra"
status
200
api_type
"completions"
id
"gen-1771615726-A4C8knyRw0YINHotICp3"
upstream_id
"chatcmpl-Rl8nLdIQS6qnJ8Lb6iRpEQdr"
total_cost
0.00002392
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepInfra"
Evaluation details
Result
Evaluator
Details
Meta Data
13.53%
Count dialogue tags
n/a
neededClean
false
dialogueTags
0
"a voice observed"
1
"Harding replied"