NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-05-10-01KR8Z9VG64DW6YD8FQEH7GJ47
Run Details
Model:
Gemma 4 26B
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
2.5s
Cost:
$0.000116
Provider:
SiliconFlow
Evaluation summary
Score
Evaluator
Details Link
43.38%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
21.6892%
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
Molly spread the blueprints across the grease-stained table. "The target is the Miller residence, specifically the backyard gazebo during the bouncy castle hour," she whispered. Bill squinted at the colorful sketches. "Wait, we are robbing a six-year-old’s birthday party? Is there even anything worth taking?" "Focus, Bill," Molly snapped. "It is about the principle and the high-end lithium-ion batteries inside those remote-controlled drones." "And the Lego sets?" Bill asked, eyes gleaming. "I hear the toddler has a limited edition Star Wars collection." Molly tapped a red circle on the map. "We strike at 1400 hours, right when the sugar crash hits. We enter through the sliding glass door while the clown is distracting the adults." "What if the clown is a professional?" Bill fretted. "A clown could be an undercover operative." "It is a man in face paint eating lukewarm pizza, Bill. Not a fed," Molly sighed. "Fine," Bill muttered, checking his lockpicks. "But I want dibs on the glitter bombs. If I'm going to jail, I want to go out fabulous." "Just grab the toys and keep your mouth shut," Molly commanded.
created_at
"2026-05-10T12:55:24.942Z"
5/10/2026, 12:55:24 PM
model
"google/gemma-4-26b-a4b-it-20260403"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
416
moderation_latency
(null)
generation_time
2297
tokens_prompt
77
tokens_completion
295
native_tokens_prompt
79
native_tokens_completion
266
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)
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
0.00011588
router
(null)
provider_responses
0
endpoint_id
"eb382d64-dce1-42d5-b23d-6fefb91948a3"
id
"chatcmpl-a050948d6a7b81b1"
is_byok
false
latency
416
model_permaslug
"google/gemma-4-26b-a4b-it-20260403"
provider_name
"SiliconFlow"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.1; linux; x64))"
http_referer
(null)
request_id
"req-1778417724-rS2vmJ3CWXS019eSlJ8o"
session_id
(null)
api_type
"completions"
id
"gen-1778417724-N7hvvqExTHw05o2sq6n5"
upstream_id
"chatcmpl-a050948d6a7b81b1"
total_cost
0.00011588
cache_discount
(null)
upstream_inference_cost
0
provider_name
"SiliconFlow"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
43.38%
Matches word count
n/a
neededClean
false
words
183
0.00%
Dialogue to Total Word Ratio
Ratio: 77.78%, Deviation: 27.78%
neededClean
false
wordsTotal
189
wordsDialogue
147
21.6892%