NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M36TH7ESF493GKJVXZ3E8772
Run Details
Model:
Grok 4.7 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
2m 2s
Cost:
$0.036813
Provider:
xAI
Evaluation summary
Score
Evaluator
Details Link
81.27%
Matches word count
(details)
19.18%
Dialogue to Total Word Ratio
(details)
50.2247%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 50% 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
The warehouse office smelled of old coffee and ambition as Molly unrolled a hand-drawn map of the suburban backyard. Bill watched skeptically, arms crossed. "Explain it to me again," he said. "Because from where I'm standing, this sounds like a one-way ticket to community service." Molly pointed at a squiggle labeled bounce house. "We don't break in, Bill. We get invited in. I sent an email pretending to be the backup entertainers. They bit. We're the clowns now." "Clowns," Bill repeated flatly. "I retired from clowns after the incident with the cream pie and the mayor's dog." "Ancient history," she dismissed. "This is different. Presents are the target. Stacked by the patio. While you wow them with balloon swords and bad jokes, I wheel our decoy wagon painted like a giant present right up and make the swap. Real gifts out, stuffed animals we bought at the dollar store in." Bill raised an eyebrow. "Dollar store animals? The kids will riot." "They'll be too busy laughing at your oversized shoes," Molly countered. "Those things are a trip hazard masterpiece. You'll fall at least twice. Comedy." He sighed but leaned in. "And the getaway? Please tell me it's not the wagon again." "It's the wagon," she confirmed cheerfully. "But we'll attach it to my bike. Pedal like mad down the alley. If pursued, we ditch it and claim we were just delivering late gifts. Who suspects clowns?" Bill couldn't suppress a snort. "Everyone suspects clowns. That's their whole brand." "Then we lean into it," Molly said, producing a red nose and honking it experimentally. The sound was pitiful. "See? Already failing upward. Parents will thank us for the free show and never check the gift tags until we're blocks away, splitting the haul of board games and gift cards." "Gift cards to where?" Bill asked, curiosity winning. "If it's toy stores, I'm keeping the Lego ones." "Fifty-fifty, as always," she agreed. "Unless you eat all the cake samples I plan to snag. Then it's sixty-forty my way." They bickered good-naturedly over shares while assembling the costumes. Bill tried on the wig, immediately regretting every life choice that led here, as Molly snapped photos for posterity and blackmail. Laughter filled the space between their words, turning tension into tandem chuckles. She demonstrated a sleight-of-hand with a wrapped box, nearly dropping it, and he caught it with a flourish that surprised them both. "Not bad for amateurs," Bill admitted. "We're not amateurs," Molly corrected. "We're specialists in the ridiculous. This heist will be talked about in thief circles for years. Or at least in our group chat." As they packed up, the plan felt less like crime and more like an elaborate prank they might actually pull off through sheer audacity and poor judgment. Bill pocketed the squeaker, Molly the map, and they headed out, already composing the excuses they'd need if a sharp-eyed eight-year-old saw through the face paint. The night air greeted them with possibility, or at least with the promise of a story too good not to tell, presents or no presents.
created_at
"2026-09-23T09:46:14.879Z"
9/23/2026, 9:46:14 AM
model
"x-ai/grok-4.7-20260916"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
577
moderation_latency
(null)
generation_time
121979
tokens_prompt
75
tokens_completion
2104
native_tokens_prompt
1303
native_tokens_completion
7555
native_tokens_completion_images
(null)
native_tokens_reasoning
6861
native_tokens_cached
1280
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
0.0368128
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790156774-QrpyTsNaOOIVabiBhezH"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790156774-mLSXqq4wtZEBe9g3FpuO"
upstream_id
"f060852e-b213-969e-913e-eea12a1f052c"
provider_responses
0
endpoint_id
"c942f37a-87e1-4fa8-83d0-8c67fda4fb4a"
id
"f060852e-b213-969e-913e-eea12a1f052c"
is_byok
false
latency
192
model_permaslug
"x-ai/grok-4.7-20260916"
provider_name
"xAI"
status
200
total_cost
0.0368128
cache_discount
0.001536
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
81.27%
Matches word count
n/a
neededClean
false
words
512
19.18%
Dialogue to Total Word Ratio
Ratio: 55.75%, Deviation: 5.75%
neededClean
false
wordsTotal
522
wordsDialogue
291
50.2247%