NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-01-01M1ENJ08DDCY9CM1Y2A145WED
Run Details
Model:
Thinking Machines Inkling Small (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
55.0s
Cost:
$0.006259
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
0.21%
Matches word count
(details)
100.00%
Dialogue to Total Word Ratio
(details)
50.1048%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 10% 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 adjusted her black beret and peered through binoculars at balloons and screaming toddlers. The target was a living room fortified by juice boxes and regret. "The cake is a fortress with sprinkles," she whispered. Bill leaned against a tree in a burglar suit that looked like a disgraced clown costume. He checked a watch made of repurposed snack coins. "And the gifts are shiny and terrifying," he agreed, adjusting a mask that smelled like cheap rubber and broken dreams. They surveyed the driveway where an angry balloon dog stood guard. Molly sketched a plan on a napkin using yogurt as ink. "We strike at seven exactly," she declared, though neither had agreed on a time zone. Bill produced a grappling hook fashioned from a bent party streamer and a paperclip. "I'll bring the rope and hope," he said, sounding more confident than his trembling hands deserved. The rope was actually a sweet-and-salty twist, but desperation made strange heroes. Together they crept toward the pastel fortress, stepping over juice boxes and the occasional crying child with amazing stealth. Molly whispered about not waking the parents, though the real danger was a finger-painting station smelling of existential dread. By the time they reached the patio sliding door, both had forgotten which one was supposed to disable the piñata alarm, which was unfortunate because the piñata was already on fire.
created_at
"2026-09-01T14:21:49.207Z"
9/1/2026, 2:21:49 PM
model
"thinkingmachines/inkling-small-20260730"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
199
moderation_latency
(null)
generation_time
54959
tokens_prompt
75
tokens_completion
4325
native_tokens_prompt
74
native_tokens_completion
5188
native_tokens_completion_images
(null)
native_tokens_reasoning
3972
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
0.0062589
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788272509-jHqB6yW5q4HZCvuH0HFc"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788272509-MEA0UZvqYlaBabVRcz6y"
upstream_id
"chatcmpl-RQ7yhRu74MPzjSBMn5il5kdq"
provider_responses
0
endpoint_id
"d1d9657b-3c3a-49b5-8f81-33204e7c71c3"
id
"chatcmpl-RQ7yhRu74MPzjSBMn5il5kdq"
is_byok
false
latency
73
model_permaslug
"thinkingmachines/inkling-small-20260730"
provider_name
"DeepInfra"
status
200
total_cost
0.0062589
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
0.21%
Matches word count
n/a
neededClean
false
words
228
100.00%
Dialogue to Total Word Ratio
Ratio: 10.82%, Deviation: 0.82%
neededClean
false
wordsTotal
231
wordsDialogue
25
50.1048%