NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHY9WR7A8CR7GEPAJZJ9XF4Q
Run Details
Model:
Grok 4 Fast
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
5.0s
Cost:
$0.000262
Provider:
xAI
Evaluation summary
Score
Evaluator
Details Link
81.27%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
40.6364%
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, blueprints of the suburban McMansion spread out like a treasure map for toddlers. Bill paced behind her, twirling a fake mustache that kept slipping off his sweaty upper lip. The target: Timmy Turner's eighth birthday bash, a piñata-packed extravaganza loaded with gadgets, games, and enough gift wrap to choke a unicorn. "Disguises first," Molly whispered, pinning a clown nose to her jacket. "We go in as party crashers—me as the balloon lady, you as the face-painter with a criminal record." Bill snorted, adjusting his oversized bow tie. "Why not just grab the gifts during the cake smash? Kids'll be too sticky to notice." "Brilliant, but timing's key," she replied, circling the backyard on the map. "We slip in at 2 PM, when Pin the Tail's in full swing. Load the van with presents while they're blindfolded." He grinned, imagining stacks of Lego sets and remote-control cars. "Escape route: bounce house to the fence. I'll rig it with hidden zip ties." Molly nodded, suppressing a giggle at the sheer ridiculousness. Stealing from snot-nosed brats felt low, but the payday? Pure gold. As sirens wailed faintly outside—probably just an ice cream truck—they clinked coffee mugs, sealing their pint-sized plot.
created_at
"2026-02-20T19:54:40.245Z"
2/20/2026, 7:54:40 PM
model
"x-ai/grok-4-fast"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1921
moderation_latency
(null)
generation_time
4874
tokens_prompt
77
tokens_completion
328
native_tokens_prompt
216
native_tokens_completion
488
native_tokens_completion_images
0
native_tokens_reasoning
206
native_tokens_cached
165
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
0.00026245
router
(null)
provider_responses
0
endpoint_id
"c9586d9d-97ff-40ca-ae94-22504d6f9b7e"
id
"8724a677-00a6-a1ab-7333-23f7ecb3c931"
is_byok
false
latency
177
model_permaslug
"x-ai/grok-4-fast"
provider_name
"xAI"
status
200
api_type
"completions"
id
"gen-1771617280-r7jLfBxIniFXfD8ng05M"
upstream_id
"8724a677-00a6-a1ab-7333-23f7ecb3c931"
total_cost
0.00026245
cache_discount
0.00002475
upstream_inference_cost
0
provider_name
"xAI"
Evaluation details
Result
Evaluator
Details
Meta Data
81.27%
Matches word count
n/a
neededClean
false
words
212
0.00%
Dialogue to Total Word Ratio
Ratio: 36.32%, Deviation: 26.32%
neededClean
false
wordsTotal
212
wordsDialogue
77
40.6364%