NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHY8147C85SJFEVD46RG0BCQ
Run Details
Model:
Grok 4 Fast
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
5.1s
Cost:
$0.000247
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 cluttered kitchen table in their dingy apartment, blueprints of the suburban backyard sprawled out like a treasure map for toddlers. Pinwheels and balloon sketches dotted the edges, marking the chaos of little Timmy's fifth birthday bash. Bill paced behind her, twirling a fake mustache he'd insisted on for "disguise purposes." "Look at this haul," Molly whispered, eyes gleaming. "Piles of shiny toys, unwrapped and waiting. We crash the gate as party clowns, snag the gift table during piñata time, and vanish in the bounce house smoke." Bill snorted, adjusting his oversized bowtie. "Clowns? With my allergies? One balloon pops, and I'm sneezing glitter everywhere." "Exactly! Cover for the chaos," she shot back. "You juggle the decoys while I stuff the sack. In and out before cake time." He leaned in, peering at the map. "What if the kids swarm us? Tiny hands grabbing loot back?" Molly grinned wickedly. "We bribe 'em with candy. Every thief's secret weapon." Bill chuckled. "Genius. Operation Sugar Rush it is." As sirens wailed faintly outside—probably just a neighbor's TV—they high-fived, oblivious to the clown wig tangled in Bill's shoelace. The heist of the century? Or a recipe for slapstick disaster? Only the party would tell.
created_at
"2026-02-20T19:22:06.454Z"
2/20/2026, 7:22:06 PM
model
"x-ai/grok-4-fast"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1556
moderation_latency
(null)
generation_time
4958
tokens_prompt
77
tokens_completion
332
native_tokens_prompt
216
native_tokens_completion
472
native_tokens_completion_images
0
native_tokens_reasoning
185
native_tokens_cached
215
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
0.00024695
router
(null)
provider_responses
0
endpoint_id
"c9586d9d-97ff-40ca-ae94-22504d6f9b7e"
id
"d20fe4bb-da81-557d-d305-4066b1aa4357"
is_byok
false
latency
110
model_permaslug
"x-ai/grok-4-fast"
provider_name
"xAI"
status
200
api_type
"completions"
id
"gen-1771615326-iINS4ZaDYgCu8YQgavCk"
upstream_id
"d20fe4bb-da81-557d-d305-4066b1aa4357"
total_cost
0.00024695
cache_discount
0.00003225
upstream_inference_cost
0
provider_name
"xAI"
Evaluation details
Result
Evaluator
Details
Meta Data
97.63%
Matches word count
n/a
neededClean
false
words
207
0.00%
Dialogue to Total Word Ratio
Ratio: 44.44%, Deviation: 34.44%
neededClean
false
wordsTotal
207
wordsDialogue
92
48.8138%