NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-12-01KKHYYN1NW7K9SZA736HQT5AP
Run Details
Model:
Grok 4.20 (Beta)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
3.8s
Cost:
$0.003956
Provider:
xAI
Evaluation summary
Score
Evaluator
Details Link
98.71%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
49.3562%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 30% 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 crouched behind the rhododendron bush, peering through the slats of the garden fence at the bouncy castle where twenty sugar-high six-year-olds screamed like tiny barbarians. “This is the worst idea you’ve ever had, Bill, and you once suggested we rob a police station with a fruit basket.” Bill adjusted his comically oversized clown wig, which kept sliding over one eye. “It’s brilliant. Kids don’t even know how to guard presents. They’re distracted by clowns and cake. We’ll be in and out before the ice cream melts. Besides, you said you wanted something low-risk after the museum fiasco.” “Low-risk? We’re about to traumatize thirty children on a Saturday afternoon. Their tiny therapy bills alone will bankrupt us.” Molly tugged at the neck of her itchy princess dress. The tiara kept stabbing her scalp like a vindictive fork. Bill grinned, revealing a gold tooth that somehow made him look even less trustworthy. “Think of it as redistribution of wealth. Those presents are just sitting there, Molly. Wrapped in shiny paper like little capitalist traps. We’re basically modern-day Robin Hoods, except with better outfits and less medieval justice.” “Robin Hood didn’t wear a purple tutu,” she muttered, adjusting the ridiculous layers of tulle around her hips. “And he certainly didn’t have to explain to a birthday boy why his new Lego set mysteriously vanished.” “Details.” Bill checked his plastic pocket watch, which had a smiling dinosaur on it. “Look, we wait for the magic show. When that terrible magician pulls the rabbit out of the hat, every kid will be staring. That’s our window. You create a diversion by pretending to be the hired entertainer’s assistant, I slip round the back and bag the presents. Simple.” Molly stared at him. “You want me to assist a children’s magician? Bill, I once punched a mime for crowding me. I’m not exactly known for my rapport with small people.” “You’ll be fine. Just smile and wave. Kids love princesses. Especially princesses who look like they’re plotting murder.” He patted her shoulder encouragingly. “Besides, I’ve got the van parked three streets over with ‘Happy Birthday Timmy’ written on the side in washable paint. We’ll blend right in.” She sighed, rubbing her temples. “If we get caught, I’m telling them you kidnapped me and forced me into this dress at gunpoint.” “Fair enough.” Bill peeked through the fence again as a fresh wave of shrieking laughter erupted. “There’s the cake coming out. Ten minutes until sugar coma. You ready, partner?” Molly looked down at her sparkling plastic shoes, then at the pile of brightly wrapped boxes visible through the patio doors. “I hate you so much right now. Let’s go steal some childhood dreams.” Bill offered her a high-five. “For the children.” She slapped his hand away. “For the cash value on eBay, you absolute clown.” They slipped through the garden gate just as the birthday song began, two thieves in ridiculous costumes moving toward their most humiliating score yet.
created_at
"2026-03-12T21:23:58.649Z"
3/12/2026, 9:23:58 PM
model
"x-ai/grok-4.20-beta-20260309"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
321
moderation_latency
(null)
generation_time
3674
tokens_prompt
77
tokens_completion
814
native_tokens_prompt
182
native_tokens_completion
637
native_tokens_completion_images
0
native_tokens_reasoning
0
native_tokens_cached
128
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
0.0039556
router
(null)
provider_responses
0
endpoint_id
"33f49f24-a0de-4dcf-872d-ca514e83fbae"
id
"173f0092-97bf-fdda-95d0-ee370755827a"
is_byok
false
latency
280
model_permaslug
"x-ai/grok-4.20-beta-20260309"
provider_name
"xAI"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.0; linux; x64))"
http_referer
(null)
api_type
"completions"
id
"gen-1773350638-VVchtB2yiCVw4RBP44Lv"
upstream_id
"173f0092-97bf-fdda-95d0-ee370755827a"
total_cost
0.0039556
cache_discount
0.0002304
upstream_inference_cost
0
provider_name
"xAI"
Evaluation details
Result
Evaluator
Details
Meta Data
98.71%
Matches word count
n/a
neededClean
false
words
494
0.00%
Dialogue to Total Word Ratio
Ratio: 63.20%, Deviation: 33.20%
neededClean
false
wordsTotal
500
wordsDialogue
316
49.3562%