NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-01-01M1ENB9WM4PW4ZJJPDKPW1T66
Run Details
Model:
Grok 4.6 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
3m 14s
Cost:
$0.065064
Provider:
xAI
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
45.95%
Dialogue to Total Word Ratio
(details)
22.9748%
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 paced the length of the small room they used as their planning headquarters, her boots clicking on the concrete floor. Bill sat quietly, flipping through a notebook filled with sketches of the target house. The children's party was scheduled for Saturday afternoon, and they had less than two days to perfect their scheme to steal all the presents. The invitation they had intercepted showed it would be a large gathering with many guests bringing gifts. Molly stopped pacing and sat down, the weight of the decision hanging in the air between them. They had never attempted something so public and so potentially disastrous, yet the allure of easy loot was strong. The warehouse around them was filled with remnants of past jobs, old tools and disguises that had served them well in more conventional thefts. "Bill, we need to decide on our disguises tonight. I think clowns are the way to go. Big shoes, red noses, the whole nine yards. We can hide bags under the costumes." Bill looked doubtful. "Clowns at a kids party might work, but I look terrible in white face paint. Maybe we should be pirates instead. Eye patches, hats. Kids like pirates." "Pirates? That's even more noticeable. No, clowns blend in with the fun atmosphere. We enter, do a little act, then start collecting 'donations' which are the presents." They talked about the sequence of events. Molly wanted to time it with the arrival of guests so they could mix in without drawing too much attention from the hosts. The house was a typical two-story suburban home with a large backyard perfect for hiding if things went south. They had scouted it the day before, noting the lack of cameras and the friendly neighborhood that wouldn't suspect a thing. "We'll arrive with the first guests, claim we were invited as surprise. Then work our way to the gift area." Bill added, "I still think we should have an escape route planned. The backyard fence is low, we can hop it if needed." "Yes, and the car parked two blocks away. We carry the presents in laundry bags, look like we're helping with cleanup." The discussion lasted well into the night as they considered every angle, from potential questions from parents to how to handle excited children who might want to play with them. They laughed at the mental image of themselves in clown makeup running down the street with armfuls of wrapped boxes, parents in pursuit. It was the kind of story that would be told in bars for years if they succeeded. Molly sketched a quick map on a napkin, marking the living room where gifts would pile up and the kitchen as a possible staging area. Bill calculated the number of trips needed based on an estimate of thirty presents, each of varying sizes. They agreed that speed was essential, no lingering to admire the haul until they were far away. They spent the remaining time gathering supplies: a cheap clown kit from the store, some sacks, and a fake invitation of their own to show if asked. Extra makeup, wigs, and even a small horn for authentic clown effect. The plan was crazy, but that's what made it perfect for their style of heist. Comedic, unexpected, and hopefully profitable. As they packed the last items, a sense of anticipation filled the room, mixed with the usual pre-job nerves.
created_at
"2026-09-01T14:18:09.692Z"
9/1/2026, 2:18:09 PM
model
"x-ai/grok-4.6-20260810"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
505
moderation_latency
(null)
generation_time
194430
tokens_prompt
75
tokens_completion
2803
native_tokens_prompt
267
native_tokens_completion
10819
native_tokens_completion_images
(null)
native_tokens_reasoning
10142
native_tokens_cached
256
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
num_fetches
(null)
web_search_engine
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"completed"
service_tier
"default"
usage
0.065064
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788272289-wvtfPpWyCl25GJMTp0mT"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788272289-ThUA1d66aeGQXb603lOQ"
upstream_id
"d5899223-2570-9ed2-b487-084f920a2783"
provider_responses
0
endpoint_id
"0d0536e3-7eb1-4acb-8249-c8813365c2d8"
id
"d5899223-2570-9ed2-b487-084f920a2783"
is_byok
false
latency
149
model_permaslug
"x-ai/grok-4.6-20260810"
provider_name
"xAI"
status
200
total_cost
0.065064
cache_discount
0.000384
upstream_inference_cost
0
provider_name
"xAI"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
567
45.95%
Dialogue to Total Word Ratio
Ratio: 26.06%, Deviation: 3.94%
neededClean
false
wordsTotal
568
wordsDialogue
148
22.9748%