NC Bench
Overview
About
Tests
Models
Vendors
Compare
Run: 2026-02-25-01KJ95XBFWGR1KDTX5A8K2FB03
Run Details
Model:
Gemini 3 Flash (Preview, Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
21.7s
Cost:
$0.012481
Provider:
Google AI Studio
Evaluation summary
Score
Evaluator
Details Link
99.92%
Matches word count
(details)
96.93%
Dialogue to Total Word Ratio
(details)
98.4235%
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 adjusted her binoculars. "The inflatable dragon is our primary extraction point," she whispered. "The bouncy one?" Bill asked. "Exactly. It provides verticality and soft-landing capabilities. Now, what’s the status on the perimeter?" "The picket fence is only three feet high, but the toddlers are the real threat. They're erratic, Molly. Low centers of gravity, sticky fingers. If one tags me, the mission is toasted." "That’s why we’re using the cake as a tactical diversion." "The chocolate fudge?" "It’s a literal sugar bomb. Once the frosting frenzy hits, we move on the gift table." "My intel says a Lego Death Star is in those boxes." "High-value loot. I like it. You got the camo?" "I’m wearing this stupid cone hat, aren't I? I feel ridiculous." "No, you look like a disgruntled uncle. It’s the perfect disguise. Did you pack the suppression gear?" "I have twelve juice boxes." "Bribes. Good. When the singing starts, that’s our window. We breach the living room, bag the haul, and hit the dragon." "What if the clown is deep-cover security?" "Gary? He’s a part-time juggler, Bill. He doesn't have the tactical training to stop us." "I don't know, Molly. His shoes are huge." Bill checked his watch. "Ready."
created_at
"2026-02-25T01:16:44.419Z"
2/25/2026, 1:16:44 AM
model
"google/gemini-3-flash-preview-20251217"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1575
moderation_latency
(null)
generation_time
21619
tokens_prompt
77
tokens_completion
1596
native_tokens_prompt
67
native_tokens_completion
4149
native_tokens_completion_images
0
native_tokens_reasoning
3824
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
0.0124805
router
(null)
provider_responses
0
endpoint_id
"6e82f652-2693-4b60-bbe0-7362d6e345f0"
is_byok
false
latency
1574
model_permaslug
"google/gemini-3-flash-preview-20251217"
provider_name
"Google AI Studio"
status
200
api_type
"completions"
id
"gen-1771982204-yLMAyEoFgbuPLThmZHqI"
upstream_id
(null)
total_cost
0.0124805
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Google AI Studio"
Evaluation details
Result
Evaluator
Details
Meta Data
99.92%
Matches word count
n/a
neededClean
false
words
203
96.93%
Dialogue to Total Word Ratio
Ratio: 94.20%, Deviation: 4.20%
neededClean
false
wordsTotal
207
wordsDialogue
195
98.4235%