NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-01-01M1EPDBNFSRDA8SF2PCN83TQP
Run Details
Model:
Grok 4.6 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
33.6s
Cost:
$0.009378
Provider:
xAI
Evaluation summary
Score
Evaluator
Details Link
0.08%
Matches word count
(details)
0.06%
Dialogue to Total Word Ratio
(details)
0.0703%
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 cramped apartment, her boots scuffing the worn linoleum as she eyed the crumpled invitation Bill had snatched from a mailbox. It was for little Timmy's sixth birthday bash, complete with a bouncy castle and enough wrapped boxes to fund their next three heists. Bill lounged on the sagging couch, twirling a fake mustache he'd bought at the dollar store, already lost in visions of glittering loot. "We dress as the catering crew," Molly said, tossing a stained apron at him. "Nobody questions the people hauling in the cake." Bill caught it and grinned. "Or we crash as balloon animals. I can twist a wiener dog that looks like a getaway car." She snorted, unfolding a napkin sketch of the backyard. Parents would be distracted by the magician, kids hyped on sugar, presents piled high like a dragon's hoard. The fence was low, the gate unlocked during peak chaos. Their van waited two blocks over, engine rumbling with impatience. Molly imagined scooping armfuls of gifts—action figures, video games, that oversized teddy that could hide smaller treasures. Bill's idea of stuffing them into hollowed-out piñatas made her laugh despite the risk. "Timing's everything," she muttered, more to herself. "Hit during the cake smash, when frosting flies and cameras flash." "And if the clown rats us out?" Bill asked, adjusting his mustache. "We bribe him with extra confetti. Or trip him into the punch bowl." They mapped escape routes over lukewarm coffee, Bill suggesting they roll under the bouncy castle like ninjas, Molly countering with a fake emergency call about a lost puppy to scatter the adults. The comedy of it all—two grown thieves tiptoeing around juice boxes and party hats—kept the tension from snapping. Molly pictured Bill's lanky frame squeezed into a unicorn costume, tripping over streamers while she pocketed the biggest boxes. He envisioned her as the stern face painter, distracting kids with lopsided rainbows while he loaded the van. Hours slipped by in bursts of planning and punches of humor. Bill practiced a magician's flourish that ended with him knocking over the lamp. Molly rehearsed a waitress smile that looked more like a snarl. They argued over the getaway: zip ties on the gate or just a polite "oops, wrong party" if spotted. The presents promised payday enough for a real vacation, not another motel flop. Molly's pulse quickened at the thought of unwrapping the haul later, laughing at the irony of grown crooks raiding a kid's joy. "Remember, no actual kids get hurt," she said firmly. "Just the wrapping paper." Bill nodded, mock-saluting. "Scout's honor. We'll leave the cake." As dawn crept in, their scheme solidified into a ridiculous masterpiece of misdirection and mayhem, two thieves ready to turn a birthday into their biggest score yet, all while dodging flying frosting and tiny accusations.
created_at
"2026-09-01T14:36:45.623Z"
9/1/2026, 2:36:45 PM
model
"x-ai/grok-4.6-20260810"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
547
moderation_latency
(null)
generation_time
33163
tokens_prompt
75
tokens_completion
914
native_tokens_prompt
267
native_tokens_completion
1506
native_tokens_completion_images
(null)
native_tokens_reasoning
878
native_tokens_cached
128
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
0.009378
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788273405-Gs8I8zCOCfwcECkrf7GR"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788273405-TZ5b4ps0RjANz0O9EvPi"
upstream_id
"8820ef17-ce42-9709-bb68-23321b11372c"
provider_responses
0
endpoint_id
"0d0536e3-7eb1-4acb-8249-c8813365c2d8"
id
"8820ef17-ce42-9709-bb68-23321b11372c"
is_byok
false
latency
145
model_permaslug
"x-ai/grok-4.6-20260810"
provider_name
"xAI"
status
200
total_cost
0.009378
cache_discount
0.000192
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
0.08%
Matches word count
n/a
neededClean
false
words
471
0.06%
Dialogue to Total Word Ratio
Ratio: 17.76%, Deviation: 12.24%
neededClean
false
wordsTotal
473
wordsDialogue
84
0.0703%