NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCYG0QT6Q6D8P0ZW3NPDJRN
Run Details
Model:
Thinking Machines Inkling
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
2m 16s
Cost:
$0.003835
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0000%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 70% 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
"If I wear the clown nose, do I have to honk?" "Molly, we are stealing every present from a fourth-grader's party, not joining the circus." "It is a circus, Bill. Bouncy castle, magician, and me in a sequined jumpsuit." "You're not in sequins. You're in black." "Black sequins. Listen, the parents think we're the balloon-arch people." "We are not balloon people." "Exactly! That's why we need the arch map. Where's the arch map?" "It's a mental map. You go left at the crying toddler." "I'm going to cry if you don't tell me the entry point." "Back gate, near the trash. The gifts are piled on a table near the piñata." "A piñata? This gets worse. What's inside?" "Probably candy and regret." "I'm taking both. How do we carry forty gifts?" "I brought garbage bags." "Classy. Do I need gloves?" "You need silence. And stop eating the sample cake." "It's buttercream, Bill. I'm a thief, not a monk. What's the exit?" "Same gate, but we sprint past the magician." "That guy sees everything. His eyes are like lasers." "Then distract him. Pull the tablecloth?" "No, too loud. I'll ask about his rabbit." "While I bag the loot?" "Exactly. You be greedy. I'll be charming." "You're never charming." "I'm charming in crisis. Once I talked a security guard into giving me a coupon." "That was a grocery store and he was your uncle." "He still counts! Okay, what's the signal?" "If I say 'the cake is a lie,' run." "Done. Do we have a code name?" "Yes. Operation Unwrap Disaster." "I hate that. Let's call it The Grinch's Revenge." "We are not green." "I'm green with envy you have the keys." "Keys?" "To the gate! You said you had them!" "I said I was getting them!" "So we scale?" "There's a hedge." "How high?" "Knee high, but thorny." "I'll wear pants." "You should wear pants regardless. It's a party, not a beach." "You're very judgmental for a man with garbage bags." "I'm practical. You're the one who brought glitter bombs." "Decoy! I throw glitter, you grab gifts. Glitter blinds adults." "Children will love you." "I'm banking on it. One kid points, I pretend to be lost." "Lost where? The parking lot?" "Lost in wonder. Bill, relax. I've done this before." "You stole a birthday cake once. You ate it in the car." "It was evidence! Okay, timeline. We arrive at three, when parents are distracted by the magician's failed card trick." "How do you know he'll fail?" "He always fails. I saw his flyer. 'Magic Mitch and His Reluctant Rabbit.'" "Reluctant? Poor bunny." "Poor us if we get caught. What's in these bags if not gifts?" "Air, mostly. And your optimism." "I have a flashlight." "Turn it off. The party has a disco ball." "Perfect. I'll dance away with the goods." "Don't dance. You're a burglar, not a DJ." "I can multitask. What's the price range on these gifts? Are we talking bicycles?" "No bikes. Board games, dolls, a drone?" "A drone? For a seven-year-old?" "Rich parents. We'll fence it by Tuesday." "To who? Your cousin with the pawn shop?" "Don't call him that. He prefers 'vintage electronics consultant.'" "He bought my toaster for two dollars." "He bought your feelings, Molly. Focus. After we grab the pile, where do we meet?" "Behind the hedge, near the garbage. Same spot." "With the loot." "And maybe a cupcake for the road." "You're impossible." "I'm hungry. Okay, I have a confession. I forgot the bags in the car." "So we use the tablecloth?" "No, I forgot the car keys too. They're on the table... inside the party." "You're kidding." "I might be. Or I might be testing your composure." "You're testing my patience." "Then we're ready. At three, I enter through the gate, you distract Magic Mitch, I bag presents, we vanish like smoke." "And if it fails?" "We blame the clown." "There is no clown." "Exactly! No alibi. Let's go steal childhood joy, Bill." "Please wear pants."<|end_message|>
created_at
"2026-07-25T15:32:28.552Z"
7/25/2026, 3:32:28 PM
model
"thinkingmachines/inkling-20260715"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
919
moderation_latency
(null)
generation_time
134033
tokens_prompt
75
tokens_completion
978
native_tokens_prompt
75
native_tokens_completion
935
native_tokens_completion_images
(null)
native_tokens_reasoning
0
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
0.00383519
router
(null)
provider_responses
0
endpoint_id
"d6b9d68e-115a-49d7-81cd-1ad1c6942b0e"
id
"chatcmpl-RX2x7EuPuqLjhCWcx8Gt4oDG"
is_byok
false
latency
171
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-1784993548-gPtGbR6dSsDJlavxJCLw"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784993548-totUzVMvnnV1jfDxLtGp"
upstream_id
"chatcmpl-RX2x7EuPuqLjhCWcx8Gt4oDG"
total_cost
0.00383519
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
0.00%
Matches word count
n/a
neededClean
false
words
664
0.00%
Dialogue to Total Word Ratio
Ratio: 100.00%, Deviation: 30.00%
neededClean
false
wordsTotal
668
wordsDialogue
668
0.0000%