NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHY8NZEQXW61DC3XF07FQ3MW
Run Details
Model:
Grok 4 Fast
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
5.7s
Cost:
$0.000249
Provider:
xAI
Evaluation summary
Score
Evaluator
Details Link
97.63%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
48.8138%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 10% 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 hunched over the rickety kitchen table, her fingers tracing a crayon-scribbled blueprint of the Peterson's backyard. Balloons bobbed in her imagination, piñatas swinging like pendulums of doom. Bill paced, his oversized trench coat dragging on the linoleum like a reluctant sidekick. "Think about it," Molly whispered, eyes gleaming. "A dozen screaming kids, distracted by cake. We swoop in as clowns—" "Clowns?" Bill snorted, stopping mid-stride. "I look ridiculous in red noses." "Exactly! Perfect cover." She jabbed the paper. "You grab the gift table while I juggle the decoys. In and out before the piñata bursts." Bill rubbed his chin, envisioning the haul: shiny bikes, glittering dolls, enough Toys R Us bounty to fence for a yacht. But the risks—sticky-fingered toddlers, hyper parents with video cams—made his stomach twist like a licorice whip. Molly leaned back, smirking. "Worst case, we blame the dog. Or the birthday boy." Bill chuckled, but his laugh echoed hollow. "Fine. But if a kid cries, you're singing 'Happy Birthday' as distraction." As dawn crept through the blinds, their plan solidified: midnight mayhem under confetti skies. Who knew larceny could feel so childishly fun?
created_at
"2026-02-20T19:33:29.702Z"
2/20/2026, 7:33:29 PM
model
"x-ai/grok-4-fast"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1314
moderation_latency
(null)
generation_time
5536
tokens_prompt
77
tokens_completion
318
native_tokens_prompt
216
native_tokens_completion
463
native_tokens_completion_images
0
native_tokens_reasoning
185
native_tokens_cached
171
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
"completed"
usage
0.00024905
router
(null)
provider_responses
0
endpoint_id
"c9586d9d-97ff-40ca-ae94-22504d6f9b7e"
id
"70c6d145-433e-d390-6870-4026bb395646"
is_byok
false
latency
78
model_permaslug
"x-ai/grok-4-fast"
provider_name
"xAI"
status
200
api_type
"completions"
id
"gen-1771616009-nlqM8wM46I8NRnOrybnT"
upstream_id
"70c6d145-433e-d390-6870-4026bb395646"
total_cost
0.00024905
cache_discount
0.00002565
upstream_inference_cost
0
provider_name
"xAI"
Evaluation details
Result
Evaluator
Details
Meta Data
97.63%
Matches word count
n/a
neededClean
false
words
193
0.00%
Dialogue to Total Word Ratio
Ratio: 33.16%, Deviation: 23.16%
neededClean
false
wordsTotal
193
wordsDialogue
64
48.8138%