NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHY8D28966X41Z973GXE2KVK
Run Details
Model:
Grok 4 Fast
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
4.4s
Cost:
$0.000270
Provider:
xAI
Evaluation summary
Score
Evaluator
Details Link
9.61%
Matches word count
(details)
70.90%
Dialogue to Total Word Ratio
(details)
40.2554%
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 hunched over the cluttered kitchen table, blueprints of the community center spread out like a pirate's treasure map. Bill paced nearby, twirling a toy sheriff's badge he'd swiped from a dollar store. The plan was ridiculous, but that's what made it brilliant—or so they told themselves. "Listen, Bill," Molly said, tapping the diagram. "We hit the party at peak chaos: piñata time. Kids'll be blindfolded and whacking away. We slip in as clowns." Bill snorted, adjusting his fake mustache. "Clowns? Molly, I look like a deranged mime in this getup. What if some snot-nosed brat recognizes me from that mall fiasco?" "Relax," she replied, grinning. "We'll use the bounce house as cover. Crawl under, snag the present pile from the back room. Easy peasy." He stopped pacing, eyes widening. "And the escape? Balloons? We float out like the Goodyear blimp?" Molly laughed. "No, genius. We load the van disguised as the ice cream truck. Honk twice, kids swarm, we peel out amid the tears." Bill rubbed his chin. "One hitch: what if the birthday kid's a tiny detective? Those rugrats are sharper than you think." She waved him off. "Then we bribe 'em with candy. It's a heist, not a horror show." As they high-fived, the room filled with gleeful scheming, oblivious to the pint-sized peril ahead.
created_at
"2026-02-20T19:28:37.652Z"
2/20/2026, 7:28:37 PM
model
"x-ai/grok-4-fast"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1736
moderation_latency
(null)
generation_time
4312
tokens_prompt
77
tokens_completion
340
native_tokens_prompt
216
native_tokens_completion
518
native_tokens_completion_images
0
native_tokens_reasoning
208
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
0.00026995
router
(null)
provider_responses
0
endpoint_id
"c9586d9d-97ff-40ca-ae94-22504d6f9b7e"
id
"6f4720fe-1b59-0e91-6e3e-42d003ac4690"
is_byok
false
latency
159
model_permaslug
"x-ai/grok-4-fast"
provider_name
"xAI"
status
200
api_type
"completions"
id
"gen-1771615717-AiLSxXv2FuzKn7DO4vCp"
upstream_id
"6f4720fe-1b59-0e91-6e3e-42d003ac4690"
total_cost
0.00026995
cache_discount
0.00003225
upstream_inference_cost
0
provider_name
"xAI"
Evaluation details
Result
Evaluator
Details
Meta Data
9.61%
Matches word count
n/a
neededClean
false
words
222
70.90%
Dialogue to Total Word Ratio
Ratio: 57.66%, Deviation: 7.66%
neededClean
false
wordsTotal
222
wordsDialogue
128
40.2554%