NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-05-10-01KR8YNH8AJHCVY9467M01E1DT
Run Details
Model:
Gemma 4 26B (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
15m 1s
Cost:
$0.000100
Provider:
DekaLLM
Evaluation summary
Score
Evaluator
Details Link
99.74%
Matches word count
(details)
99.92%
Dialogue to Total Word Ratio
(details)
99.8343%
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
The air was heavy with tension. Molly slammed a crayon-drawn map onto the scarred kitchen table, her expression grim and intensely focused. "The perimeter is crawling with sticky-handed operatives, Bill. It’s a high-sugar nightmare." Bill squinted at the colorful, messy scribbles, leaning back heavily in his creaky wooden chair while sighing deeply. "Are you seriously worried about five-year-olds?" "The toddlers are unpredictable," Molly hissed, her eyes narrowing into dark, suspicious slits. "They have no concept of personal space or stealth." "What about the clown?" Bill asked, nervously picking at a loose thread on his fraying sleeve. "The clown is a chaotic neutral variable. We move during the sugar crash." "And the parents?" Bill inquired, glancing tentatively toward the darkened window. "Juice boxes provide a distraction," Molly explained, pointing at a drawing of a carton. "Target the LEGO sets and remote-control cars. Grab the heavy-duty sacks." Bill crossed his arms over his chest. "I am not wearing a glittery party hat, Molly. I have standards." "You’ll wear glitter if you want those dinosaur figurines," Molly countered, staring him down with iron resolve. "Check your bag for lockpicks." "It’s a tactical satchel!" Bill protested, standing up abruptly from the table. "Whatever. Just get the juice boxes."
created_at
"2026-05-10T12:44:19.089Z"
5/10/2026, 12:44:19 PM
model
"google/gemma-4-26b-a4b-it-20260403"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
8958
moderation_latency
(null)
generation_time
901310
tokens_prompt
77
tokens_completion
12785
native_tokens_prompt
68
native_tokens_completion
290
native_tokens_completion_images
(null)
native_tokens_reasoning
13175
native_tokens_cached
0
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
num_fetches
(null)
web_search_engine
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
service_tier
(null)
usage
0.00009978
router
(null)
provider_responses
0
endpoint_id
"d99bcbe6-bec3-4991-965e-45fea79fe63c"
id
"chatcmpl-a38380d11a8b9989"
is_byok
false
latency
8958
model_permaslug
"google/gemma-4-26b-a4b-it-20260403"
provider_name
"DekaLLM"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.1; linux; x64))"
http_referer
(null)
request_id
"req-1778417059-4dOd0UhcujZuHFxovwJV"
session_id
(null)
api_type
"completions"
id
"gen-1778417059-8Rm4YMzJR6UQiFWDwfUk"
upstream_id
"chatcmpl-a38380d11a8b9989"
total_cost
0.00009978
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DekaLLM"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
99.74%
Matches word count
n/a
neededClean
false
words
204
99.92%
Dialogue to Total Word Ratio
Ratio: 51.66%, Deviation: 1.66%
neededClean
false
wordsTotal
211
wordsDialogue
109
99.8343%