NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M37HZJFDVG0H166M7CVDP392
Run Details
Model:
GPT-6 Sol (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
19.0s
Cost:
$0.008952
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
4.93%
Dialogue to Total Word Ratio
(details)
2.4668%
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 unfolded a floor plan across the café table. It was drawn in crayon, with a large purple square labeled DOOM. Bill squinted. “Is that the party venue?” “That’s the bouncy castle.” “Why does it say doom?” “Because you’re carrying the presents.” Bill studied the drawing. “Where’s the front door?” “Under the spaghetti stain. The party starts at two. We enter at two fifteen, when the children are distracted by the magician.” “And the parents?” “Distracted by the magician’s invoice.” Bill tapped a red circle. “What’s this?” “Present table. Our target.” “It’s awfully close to the cake.” “You’re afraid of cake?” “I’m afraid of children near cake. Their elbows go everywhere.” Molly pulled a packet of balloons from her coat. “That’s why we’re going in as entertainers. I’m a balloon artist. You’re my assistant.” “Can you make balloon animals?” “I can make a snake.” “That’s just an inflated balloon.” “It’s an exceptionally convincing snake.” A waitress set down two coffees. Molly waited until she’d gone, then produced a tiny silver van from her pocket. Bill picked it up. “Why have you brought a toy van?” “Visual aid. We park here, beside the hedge. We load the presents through the kitchen window, drive away, and split the proceeds.” “How big is the actual van?” “About the same proportions, only larger.” “That’s reassuring. How many presents are there?” “Thirty-seven, according to the invitation photograph.” “You counted them?” “Twice. The pile grew by three between photos. Someone’s aunt is overcompensating.” Bill rubbed his forehead. “We’ll need sacks.” “Already acquired.” “Disguises?” “Already acquired.” “An explanation for why your balloon artist has sacks?” Molly paused. “Professional balloon storage.” “And when they see us putting gift-wrapped boxes in the storage?” “Those are the balloons’ personal effects.” Bill slid the toy van back. “I’d like to renegotiate my fee.” “You get half.” “I’d like half plus dignity.” “Dignity isn’t in the budget.” “What are the presents worth, anyway?” “Unknown. The birthday child is six. Potentially expensive building sets.” “Potentially socks.” “Then we become wealthy in socks.” Molly flipped the floor plan over. On the back was a schedule titled OPERATION HAPPY BIRTHDAY, with arrows crossing out several increasingly elaborate schemes. Bill read one. “Why is ‘hide inside piñata’ crossed out?” “Weight limit.” “And ‘train squirrels’?” “Union dispute.” He looked up. “Molly, how are we getting past the children?” “We tell them we’re running a game.” “What game?” “Find the missing presents.” “That seems likely to draw attention to the missing presents.” “Not if we tell them to count to a hundred first.” “They’re six. One of them will reach a hundred by saying ‘ten’ repeatedly.” Molly folded the plan. “Fine. New approach. We don’t sneak in. We walk straight up to the birthday child and ask where the presents go.” “Why would that work?” “Because everyone at a children’s party is carrying something, and nobody knows where it belongs.” Bill considered this. “What if someone asks who hired us?” “We say the other parent.” “And if there’s only one parent?” “We say the magician.” “You really don’t like that magician.” “I saw his invoice.” Bill reached for his coffee. “One last question. Why steal all the presents?” Molly stared at him. “Because the plan says all.” “That’s not an answer.” “It is if you wrote the plan in crayon.”
created_at
"2026-09-23T16:36:02.17Z"
9/23/2026, 4:36:02 PM
model
"openai/gpt-6-sol-20260922"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
5456
moderation_latency
(null)
generation_time
18944
tokens_prompt
75
tokens_completion
1144
native_tokens_prompt
66
native_tokens_completion
882
native_tokens_completion_images
0
native_tokens_reasoning
100
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
0.008952
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790181362-d7BNoQs2E86Tf3nGFI0l"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790181362-faMvVeW6gELOiicufUk0"
upstream_id
"resp_00cea7f537fee8b7016ab3fff24e7887d19fe33e27e0c44b64"
provider_responses
0
endpoint_id
"1cfc7d9d-4404-4b8e-9ee4-58ae45c9dcd4"
id
"resp_00cea7f537fee8b7016ab3fff24e7887d19fe33e27e0c44b64"
is_byok
false
latency
2968
model_permaslug
"openai/gpt-6-sol-20260922"
provider_name
"OpenAI"
status
200
total_cost
0.008952
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
552
4.93%
Dialogue to Total Word Ratio
Ratio: 77.76%, Deviation: 7.76%
neededClean
false
wordsTotal
553
wordsDialogue
430
2.4668%