NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-12-01KKHZ9F3GKMPXW76TD2V6AXA2
Run Details
Model:
Grok 4.20 (Beta)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
1.8s
Cost:
$0.001730
Provider:
xAI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
49.9995%
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, you’re kidding me. A children’s party? We’re stealing from toddlers now? Keep your voice down, Bill. It’s not toddlers, it’s their presents. Twenty little brats, one giant pile of wrapped joy in the middle of the rec center. In and out in ninety seconds. You want me to jimmy a window while some sugar-high kid is pelting me with cake? We go in as the hired magicians. I’m the beautiful assistant, you’re the bumbling clown. While I dazzle them with card tricks, you wheel the present table straight out the side door like it’s part of the act. What if the birthday boy starts screaming? Then you honk your stupid clown nose and yell “Magic!” Problem solved. This is the stupidest plan we’ve ever had, and we once stole a mayor’s toupee. Exactly. Low expectations mean low security. They’ll have one bored dad by the door and a piñata full of Tootsie Rolls. We walk out millionaires in tiny socks and plastic trucks. I hate how much sense you’re making. Good. Now practice your goofy laugh. We strike at two o’clock sharp, right after “Happy Birthday.” Fine. But if I get hit with a juice box, I’m keeping the Xbox.
created_at
"2026-03-12T21:29:53.013Z"
3/12/2026, 9:29:53 PM
model
"x-ai/grok-4.20-beta-20260309"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
281
moderation_latency
(null)
generation_time
1720
tokens_prompt
77
tokens_completion
313
native_tokens_prompt
182
native_tokens_completion
266
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
0.0017296
router
(null)
provider_responses
0
endpoint_id
"33f49f24-a0de-4dcf-872d-ca514e83fbae"
id
"7a8f9ce9-e7fa-0c2f-8cc5-53a744b4ab53"
is_byok
false
latency
241
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-1773350993-wMyRad24DhYC89hlNz50"
upstream_id
"7a8f9ce9-e7fa-0c2f-8cc5-53a744b4ab53"
total_cost
0.0017296
cache_discount
0.0002304
upstream_inference_cost
0
provider_name
"xAI"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
201
0.00%
Dialogue to Total Word Ratio
Ratio: 1.49%, Deviation: 88.51%
neededClean
false
wordsTotal
202
wordsDialogue
3
49.9995%