NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-01-01M1EMNEZDHTJ5M6CEPP6BVANE
Run Details
Model:
Thinking Machines Inkling Small (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
11.5s
Cost:
$0.001874
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
69.44%
Dialogue to Total Word Ratio
(details)
34.7205%
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 leaned against the fence. "Bill, I’ve reconnoitered the venue. They’ve got a slide, a sandbox, and a woman in yoga pants who looks like she could bench-press a car." Bill checked his watch. "I prefer entry through the gift bag. It’s softer." Molly sighed. "We need to steal the presents, not build a fort out of them. The party starts in twenty minutes, and every parent is currently arguing about whether the juice is organic or just expensive." Bill nodded. "Organic juice implies moral superiority. I can work with that. What’s the heist architecture?" Molly pulled a crumpled map from her hoodie. "We enter through the bounce house. It’s structurally irrelevant, which is perfect. Once inside, we blend with the chaos. I’ll wear a tutu and claim I’m a dance instructor." Bill frowned. "You look like a burglar in a tutu." Molly shrugged. "You look like a burglar who forgot his own name. Anyway, at 3:15 all the adults will be outside arguing about sunscreen while the children activate their gift-wrecking protocols." Bill pointed at a bright pink table. "That’s the treasure. Unicorn backpacks, a tablet that plays only farm sounds, and a gift card to a toy store." Molly grinned. "Exactly. We grab the high-value items, avoid anything with a microphone, and exit through the balloon arch before they notice the absence of presents." Bill checked his pockets. "I have a crowbar, a pair of sunglasses, and a half-eaten granola bar." Molly rolled her eyes. "You’re a professional." Bill smiled. "Only after hours. Ready to ruin a child’s birthday?" Molly adjusted her tutu. "I’ve never been more ready. Let’s go before some dad starts filming us for social media." Bill followed her toward the bounce house. "Social media is the real enemy here." Molly whispered, "Remember, if we’re caught, we’re not thieves. We’re lost relatives who brought cupcakes. And your name is Gary." Bill nodded. "Gary. I hate Gary. Let’s move before Gary gets a crown."
created_at
"2026-09-01T14:06:14.006Z"
9/1/2026, 2:06:14 PM
model
"thinkingmachines/inkling-small-20260730"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
184
moderation_latency
(null)
generation_time
11429
tokens_prompt
75
tokens_completion
1668
native_tokens_prompt
74
native_tokens_completion
1534
native_tokens_completion_images
(null)
native_tokens_reasoning
1155
native_tokens_cached
0
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
"stop"
service_tier
(null)
usage
0.0018741
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788271574-yPZSabnL95j8UmqYlg6t"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788271574-arqN7du05Z20b1EemiyZ"
upstream_id
"chatcmpl-RhC6s7UwCSwBGgAYdTVFh82f"
provider_responses
0
endpoint_id
"d1d9657b-3c3a-49b5-8f81-33204e7c71c3"
id
"chatcmpl-RhC6s7UwCSwBGgAYdTVFh82f"
is_byok
false
latency
77
model_permaslug
"thinkingmachines/inkling-small-20260730"
provider_name
"DeepInfra"
status
200
total_cost
0.0018741
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepInfra"
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
328
69.44%
Dialogue to Total Word Ratio
Ratio: 82.23%, Deviation: 7.77%
neededClean
false
wordsTotal
332
wordsDialogue
273
34.7205%