NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-09-01KX40XYWRV1Y171VERQ8646JX
Run Details
Model:
GPT-5.6 Sol
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
19.6s
Cost:
$0.027990
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.02%
Dialogue to Total Word Ratio
(details)
0.0122%
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 spread the floor plan across the café table. A blob of jam marked the ballroom. “That’s the target,” she said. “That’s a scone,” Bill said. “The target is beneath the scone.” “Then our first obstacle is pastry.” “The presents arrive at two. Party begins at three. Magician at three-fifteen. Cake at four. We enter during the rabbit trick.” “Why?” “Everyone watches the hat.” “I always watch the rabbit. They look furious.” “Focus, Bill. Thirty children, twelve adults, one entertainer, and a mountain of gifts.” “Any guards?” “Two grandmothers.” Bill lowered his voice. “Professional grandmothers?” “The worst kind. Knitters.” “They’ll have needles.” “And hard sweets. Stay alert.” “How do we get inside?” Molly produced two laminated cards. “Children’s entertainers.” Bill read his. “Bongo and Marzipan’s Educational Insect Experience.” “You’re Bongo.” “I’m always Bongo.” “You have the face for it.” “What does Marzipan do?” “Marzipan carries the bags.” “I thought we were stealing presents.” “We carry them out in educational insect bags.” “What’s our insect?” “Beetles.” “Live?” “Plastic.” “Thank goodness. I was bitten by a ladybird once.” “No, you weren’t.” “It landed on me aggressively.” Molly tapped the plan. “We enter through the conservatory. You release the beetles. I locate the gifts.” “Release them where?” “Everywhere.” “Won’t the children pick them up?” “Exactly. Confusion.” “Children like beetles?” “Children like anything adults are shouting about.” A waiter approached, saw the diagrams, fake moustaches, and tub of plastic bugs, then retreated. Bill leaned closer. “Suppose the magician recognizes us?” “Why would he?” “We robbed him last Easter.” “We did not rob him. We borrowed his van.” “We returned it in a lake.” “Near a lake.” “In the shallow end of a lake.” “The point is, he never saw our faces.” “He saw mine.” “You were wearing a duck mask.” “He may recognize my posture.” “No one recognizes posture.” “My posture is distinctive. A physiotherapist once called it criminal.” Molly unfolded another sheet. “Escape route: ballroom, corridor, kitchen, loading bay.” “Kitchen? There’ll be sausage rolls.” “Do not stop.” “I’ll need energy.” “Steal first, snack later.” “That philosophy caused the Antwerp fainting.” “You ate nine éclairs.” “For energy.” “You fainted from custard.” Bill inspected the plan. “What if the gifts are labeled?” “They will be.” “That’s rather personal.” “We’re thieves.” “We can still have standards.” “Our standard is unopened.” “What if one contains a puppy?” “We leave it.” “A hamster?” “We leave it.” “A drum kit?” “We take it, but quietly.” “A chemistry set?” “Carefully.” “A book?” Molly stared at him. “What?” “You want to leave the books?” “They’re heavy.” “They’re valuable.” “To children?” “To literacy.” “We’re stealing their birthday presents, Molly. Let’s not pretend we’re librarians.” A gust from the café door lifted the floor plan. Bill trapped it with the scone. “Fine,” Molly said. “We take everything except living animals.” “And cursed objects.” “How will we know?” “They usually whisper.” “At four-ten, we’re gone.” “With the loot.” “With the educational resources.” “Presents.” “Inventory.” “Stolen presents.” “Acquired inventory.” Bill nodded. “And nobody gets hurt?” “Nobody.” “Except emotionally.” “Temporary disappointment builds character.” “That’s cold.” “You’re the one who suggested stealing the cake.” “For energy.” “We are not stealing the cake.” “Then I want hazard pay.” “You’re getting half.” “Half the presents?” “Half the proceeds.” “What if I want a present?” “You are forty-three.” “That’s not a no.” Molly gathered the papers. “Wear cheerful trousers.” “My trousers are naturally pessimistic.” “Then buy balloons.” “What do balloons have to do with beetles?” “Nothing. That’s why they’ll trust us.” Bill pocketed the plastic insects. One fell into his coffee. He peered down. “Problem?” Molly asked. “The beetles have breached containment.” “They’re plastic.” “That’s what they want us to think.”
created_at
"2026-07-09T18:06:09.319Z"
7/9/2026, 6:06:09 PM
model
"openai/gpt-5.6-sol-20260709"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1565
moderation_latency
(null)
generation_time
19523
tokens_prompt
77
tokens_completion
1417
native_tokens_prompt
66
native_tokens_completion
922
native_tokens_completion_images
0
native_tokens_reasoning
0
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
0.02799
router
(null)
provider_responses
0
endpoint_id
"a54c5de0-89bf-4ad7-a212-cf977eed918a"
id
"resp_0018225ad9127be5016a4fe31165e481979b06854115891571"
is_byok
false
latency
283
model_permaslug
"openai/gpt-5.6-sol-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-1783620369-jKZbyoyFLcaD6omvJax2"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1783620369-XR7xqGT6Mv81phC8ZqK9"
upstream_id
"resp_0018225ad9127be5016a4fe31165e481979b06854115891571"
total_cost
0.02799
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
611
0.02%
Dialogue to Total Word Ratio
Ratio: 82.90%, Deviation: 12.90%
neededClean
false
wordsTotal
614
wordsDialogue
509
0.0122%