NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-09-01KX3ZZB9A9V3AKFZ8QQ443J1F
Run Details
Model:
GPT-5.6 Terra (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
12.3s
Cost:
$0.013335
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
2.32%
Dialogue to Total Word Ratio
(details)
1.1621%
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
Molly peered over the garden fence. “There. Pink balloons, a rented pony, and at least forty gift bags. We go in at three fifteen, during the magician.” Bill adjusted his black wool cap. “Why during the magician?” “Because children become spiritually unavailable when a man pulls scarves out of his ears.” “I thought the plan was to steal presents, not insult a professional entertainer.” “Bill, he’s called Mystico the Magnificent and he’s wearing a cape with moons on it. Insult is the only language he speaks.” Bill unfolded a crumpled map. “Front gate is guarded by a clown.” “That’s not a guard.” “He’s enormous.” “He’s on stilts.” “Exactly. Elevated surveillance.” Molly sighed. “We enter through the side gate. You distract the adults.” “How?” “Pretend you’re the birthday boy’s uncle.” “I don’t know the birthday boy.” “Neither do they. Look at them. They’re all holding paper plates and arguing about gluten. Tell them you’re Uncle Trevor from Norwich.” “My name is Bill.” “Not today. Today you’re Trevor, recently divorced, emotionally fragile, and carrying a tray of sausage rolls.” Bill looked toward the garden. “Where do I get sausage rolls?” “Mrs. Pemberton’s kitchen. Her back door is open.” “That sounds like burglary.” “We’re planning to steal a mountain of presents from seven-year-olds, Bill. Let’s not get precious about the sausage rolls.” A burst of shrieking rose from the garden. Bill flinched. “What if the children notice us?” “They won’t. Children only notice two things: sugar and anyone holding an iPad.” “I don’t have an iPad.” “You have a phone.” “It’s got a cracked screen.” “Perfect. Tell them it’s a treasure map.” Bill squinted at the pile of presents beneath a gazebo. “There are bicycles.” “Take the small ones.” “There’s a dollhouse.” “Take the furniture. Less conspicuous.” “A telescope?” “Definitely. We can use it to check whether anyone’s following us.” Bill stared at her. “You really thought this through.” “I have standards.” “You once stole a mayor’s ceremonial chain because you thought it was a gold necklace.” “It was an easy mistake. It had a medallion.” “It said ‘MAYOR OF HERTFORDSHIRE.’” “In tiny letters.” The magician’s voice boomed from the garden. “And now, children, watch closely as I make this rabbit disappear!” Molly snapped her fingers. “That’s our cue.” Bill grabbed her sleeve. “Wait. We need code names.” “No, we don’t.” “All proper heists have code names.” “Fine. You’re Sausage Roll.” “Why am I Sausage Roll?” “Because you’re going to be holding them.” “And you?” “Black Widow.” “That’s quite dramatic.” “I’m wearing black.” “You’re wearing a cardigan with ducks on it.” “Black ducks.” Bill swallowed. “What if the adults ask me who Trevor is?” “Say you met the family at a mindfulness retreat.” “I’ve never been to a mindfulness retreat.” “Neither have they. That’s why it works.” They slipped through the side gate. Bill returned moments later carrying a tray of sausage rolls and wearing a party hat. Molly whispered, “Why are you wearing that?” “A child put it on me.” “Did you resist?” “He had jam on his hands.” From the lawn, a small girl pointed at Bill. “Mummy! Uncle Trevor’s here!” Bill froze. Molly grinned. “Congratulations. You’re family now.” The girl ran toward him. “Did you bring my present?” Bill looked at Molly. Molly looked at the gift pile. “Of course,” he said weakly. “I brought everybody’s.”
created_at
"2026-07-09T17:49:26.208Z"
7/9/2026, 5:49:26 PM
model
"openai/gpt-5.6-terra-20260709"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1404
moderation_latency
(null)
generation_time
12155
tokens_prompt
77
tokens_completion
1210
native_tokens_prompt
66
native_tokens_completion
878
native_tokens_completion_images
0
native_tokens_reasoning
77
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
"completed"
service_tier
"default"
usage
0.013335
router
(null)
provider_responses
0
endpoint_id
"3cc89000-ae34-4dde-9c7d-5ce009c8af0b"
id
"resp_08279866506026ce016a4fdf2647fc819088788144ddec3abd"
is_byok
false
latency
234
model_permaslug
"openai/gpt-5.6-terra-20260709"
provider_name
"OpenAI"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.18.0; linux; x64))"
http_referer
(null)
request_id
"req-1783619366-nv6Ja4JGf8j1M5GYEtBi"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1783619366-g7hzpraXbfwrdurAc1AV"
upstream_id
"resp_08279866506026ce016a4fdf2647fc819088788144ddec3abd"
total_cost
0.013335
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
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
556
2.32%
Dialogue to Total Word Ratio
Ratio: 78.67%, Deviation: 8.67%
neededClean
false
wordsTotal
558
wordsDialogue
439
1.1621%