NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCVSAH9ZCK1GNG2DWH8DW3A
Run Details
Model:
Thinking Machines Inkling (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
7m 23s
Cost:
$0.024502
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
27.17%
Matches word count
(details)
99.66%
Dialogue to Total Word Ratio
(details)
63.4143%
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
"Balloon arch or back gate?" Bill asked. "Back gate," Molly said. "The arch is decoy-obvious." "It's a five-year-old's birthday party." "Exactly. Kids have security." Bill adjusted his black turtleneck. "We blend in how?" "I'm the magician's assistant. You're the guy who smells like regret." "I smell like tactical wool." "Same difference. We go at cake time. The parents are sugar-blind." Bill nodded. "Present table?" "Under the unicorn piƱata. Which, by the way, is armed." "Armed?" "Candy and trauma." "How many gifts?" "Forty-seven minimum." "For one child?" "Name's Archduke something." "We need sacks." "Elegant sackcloth." "Gloves?" "Fingerless. I need dexterity for ribbon cutting." "Ribbon cutting?" "You never know what's wrapped in gold paper versus emotional damage." Bill sighed. "Exit?" "Same gate. I distract them with a card trick involving a missing quarter." "That's not magic. That's theft." "At this party, Bill, it's identical." "Timing?" "When the clown cries. 3:15 sharp. I've seen the schedule." "You scouted his emotional breakdown?" "I scouted everything. There's a dog in a tutu guarding the east wing." "East wing?" "It's a ranch house. Work with me." "I'm working, I'm working." "Good. Don't trip the bubble machine." "Bubble machines don't have alarms." "These do. Very wealthy child." Bill pulled on gloves. "Let's steal Christmas in July." "It's a dinosaur theme, Bill. Let's steal the Jurassic period."
created_at
"2026-07-25T14:45:06.231Z"
7/25/2026, 2:45:06 PM
model
"thinkingmachines/inkling-20260715"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
850
moderation_latency
(null)
generation_time
442837
tokens_prompt
75
tokens_completion
4631
native_tokens_prompt
75
native_tokens_completion
6038
native_tokens_completion_images
(null)
native_tokens_reasoning
4243
native_tokens_cached
32
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
0.02450234
router
(null)
provider_responses
0
endpoint_id
"d6b9d68e-115a-49d7-81cd-1ad1c6942b0e"
id
"chatcmpl-R4XnOdNXkDl9VNdxZugIe8Pl"
is_byok
false
latency
70
model_permaslug
"thinkingmachines/inkling-20260715"
provider_name
"DeepInfra"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.18.0; linux; x64))"
http_referer
(null)
request_id
"req-1784990706-IIh4RCJ85KYI2lDj4XBW"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784990706-YbMNMaoY1XIfAtmkfG3m"
upstream_id
"chatcmpl-R4XnOdNXkDl9VNdxZugIe8Pl"
total_cost
0.02450234
cache_discount
0.00002656
upstream_inference_cost
0
provider_name
"DeepInfra"
response_cache_source_id
(null)
data_region
"global"
Evaluation details
Result
Evaluator
Details
Meta Data
27.17%
Matches word count
n/a
neededClean
false
words
219
99.66%
Dialogue to Total Word Ratio
Ratio: 92.41%, Deviation: 2.41%
neededClean
false
wordsTotal
224
wordsDialogue
207
63.4143%