NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-01-01M1EN6N0XWVWFMZQNDKM5R773
Run Details
Model:
Grok 4.6 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
2m 3s
Cost:
$0.031692
Provider:
xAI
Evaluation summary
Score
Evaluator
Details Link
97.63%
Matches word count
(details)
2.10%
Dialogue to Total Word Ratio
(details)
49.8658%
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 dim warehouse, blueprints of the suburban house scattered across the crate they used as a table. Bill sat opposite, twisting a balloon into something that resembled a deflated sausage more than a dog. The air smelled of stale coffee and greasepaint from the costumes piled in the corner. "This Johnson kid's party is gold," Molly said, stabbing a finger at the photo of wrapped boxes stacked high in the living room. "Parents went all out. Tablets, bikes, cash cards. We hit it at two when the sugar rush peaks." Bill popped the balloon with a sigh. "Sugar rush means screaming. Last time we did a family thing I ended up with frosting in my ear. Remember the wedding?" "Different gig. This time we go in as the replacement clowns. I already phoned the mom, told her the agency mixed up the booking. She sounded desperate. We'll roll up in the van, honk the horn, start the show." They leaned in closer. Molly sketched arrows on the map showing the backyard fence and the side door that led straight to the gift table. Bill suggested parking two streets over so no one noticed the getaway vehicle too soon. "I'll do the magic that always fails," Molly continued. "Coins disappear into my sleeve, scarves knot themselves. Kids love a disaster. You handle balloons and face paint. Keep every parent watching the mess we make." Bill grinned, already imagining the chaos. "And when they drag out the cake I slip behind the table, dump the presents into the prop sack. Twenty seconds, tops. We vanish during the singing." Molly nodded, eyes bright. "Exactly. Leave a note that says the toys went to kids who actually need them. Makes us look like rogues with hearts. Then we pawn the electronics and split the rest." They argued over details for another hour. Bill worried about a nosy uncle with a phone. Molly waved it off, saying the oversized shoes and red noses would hide them better than any mask. She demonstrated a pratfall that sent her sprawling, drawing a genuine laugh from Bill. "See? Comedy is our cover," she said, climbing back up. "If anyone questions us we just honk louder and trip again." Bill practiced walking in the giant shoes, knocking over a stack of empty pizza boxes. Molly caught him before he hit the floor, both of them howling. The plan felt airtight in that moment, ridiculous enough to work. They packed the duffel with extra balloons and a whoopee cushion for backup diversion, then toasted with lukewarm soda. Dawn crept through the grimy windows as they finalized the route. Molly folded the map, Bill stuffed the last balloon into his pocket. Tomorrow they would crash a five-year-old's birthday, steal every present in sight, and drive off laughing at the absurdity of it all. For two career thieves who specialized in jobs that always went sideways, this one promised the best punchline yet.
created_at
"2026-09-01T14:15:37.257Z"
9/1/2026, 2:15:37 PM
model
"x-ai/grok-4.6-20260810"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
485
moderation_latency
(null)
generation_time
122799
tokens_prompt
75
tokens_completion
1701
native_tokens_prompt
267
native_tokens_completion
5257
native_tokens_completion_images
(null)
native_tokens_reasoning
4641
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
0.031692
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788272137-sKVIvi9memmALDjgRsia"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788272137-vshlf2HYcQsJlhobOZFH"
upstream_id
"2fa8ada9-facf-9734-8495-389671324ac3"
provider_responses
0
endpoint_id
"0d0536e3-7eb1-4acb-8249-c8813365c2d8"
id
"2fa8ada9-facf-9734-8495-389671324ac3"
is_byok
false
latency
170
model_permaslug
"x-ai/grok-4.6-20260810"
provider_name
"xAI"
status
200
total_cost
0.031692
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
97.63%
Matches word count
n/a
neededClean
false
words
493
2.10%
Dialogue to Total Word Ratio
Ratio: 38.79%, Deviation: 8.79%
neededClean
false
wordsTotal
495
wordsDialogue
192
49.8658%