NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHYAX73WZRJFCF08C5193HHH
Run Details
Model:
Grok 4
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
16.3s
Cost:
$0.008375
Provider:
xAI
Evaluation summary
Score
Evaluator
Details Link
93.65%
Matches word count
(details)
89.37%
Dialogue to Total Word Ratio
(details)
91.5116%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 90% 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 leaned against the dimly lit warehouse wall, her eyes sparkling with mischief as she unfolded a crumpled blueprint. Bill paced nearby, cracking his knuckles. "Alright, Bill, listen up. The target's a kiddie birthday bash at the mayor's mansion—balloons, cake, and a mountain of wrapped goodies under that piñata tree. We hit it during the clown show. You dress as a magician's assistant, all sparkly tights and distractions." Bill snorted, adjusting his fake mustache. "Sparkly tights? Molly, I'm a master thief, not a disco ball! Why not you? You've got the legs for it." "Because I'm the brains, dummy. I'll sneak in as the caterer—poison the punch with sleepy juice, non-lethal stuff. Kids nap, parents panic, we grab the gifts. Easy peasy. That diamond-encrusted Barbie alone is worth a fortune on the black market." "Sleepy juice? What if a kid's allergic? We're crooks, not monsters!" "Relax, it's just chamomile tea concentrate. They'll dream of unicorns. You handle the getaway van—park it as a bouncy house delivery. Honk like a goose if things go south." Bill grinned. "Honk like a goose? You're twisted, Mol. Fine, but if I trip over a tricycle, you're buying me new knees." "Deal. Let's make those brats' party our payday!"
created_at
"2026-02-20T20:12:24.07Z"
2/20/2026, 8:12:24 PM
model
"x-ai/grok-4-07-09"
4/7/2009, 12:00:00 AM
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
5978
moderation_latency
(null)
generation_time
16258
tokens_prompt
77
tokens_completion
326
native_tokens_prompt
744
native_tokens_completion
521
native_tokens_completion_images
0
native_tokens_reasoning
229
native_tokens_cached
743
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
0.00837525
router
(null)
provider_responses
0
endpoint_id
"d40bef9d-6936-4c57-9749-bf7399de4a77"
id
"ddfe28b1-66cb-14dc-e781-227f09bc56e8"
is_byok
false
latency
221
model_permaslug
"x-ai/grok-4-07-09"
4/7/2009, 12:00:00 AM
provider_name
"xAI"
status
200
api_type
"completions"
id
"gen-1771618344-JGgWeWIRcQhRhX4iI8RW"
upstream_id
"ddfe28b1-66cb-14dc-e781-227f09bc56e8"
total_cost
0.00837525
cache_discount
0.00167175
upstream_inference_cost
0
provider_name
"xAI"
Evaluation details
Result
Evaluator
Details
Meta Data
93.65%
Matches word count
n/a
neededClean
false
words
209
89.37%
Dialogue to Total Word Ratio
Ratio: 84.21%, Deviation: 5.79%
neededClean
false
wordsTotal
209
wordsDialogue
176
91.5116%