NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M37HX4GG09ZXSR55EF36CKYM
Run Details
Model:
GPT-6 Sol
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
15.8s
Cost:
$0.007212
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
20.19%
Matches word count
(details)
2.00%
Dialogue to Total Word Ratio
(details)
11.0941%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 50% 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 a crayon drawing across the café table. It showed a house, a cake, and a stick figure labeled MUMMY. A purple arrow pointed at the pile of presents. “Our informant is six,” Bill said. “Six and a half. Show some respect.” “He’s drawn the dog as a helicopter.” “That’s because the dog is named Chopper. Honestly, Bill, try reading the intelligence before criticizing it.” Rain ticked against the window. Across the street, balloons bobbed above the gate of number twelve. A banner announced HAPPY BIRTHDAY, ALFIE! Beneath it stood Chopper, an enormous golden retriever wearing a party hat. Bill folded his arms. “We walk through the gate, distract the adults, and take the presents?” “That is the broad strategy.” “It’s also the entire strategy.” Molly tapped the drawing. “I’ll arrive as the magician. You’ll be my assistant.” “I can’t do magic.” “Perfect. Neither can most magicians. You just need to look astonished.” “I always look astonished when you make plans.” “Excellent. When I shout ‘Behold,’ you pull the tablecloth.” “Off the table?” “No, from my bag. We drape it over the gifts, then carry everything out.” Bill looked at the sketch. “The gifts are in the living room.” “Yes.” “And the party is in the garden.” “Yes.” “So who are we performing for?” “The dog, if things go badly.” Inside number twelve, someone blew a party horn with the sustained urgency of a sinking ship. Molly adjusted her glittery top hat. It had once belonged to a pantomime villain and still smelled faintly of gravy. Bill checked his reflection in a spoon. His assistant’s costume was a waistcoat sewn with silver stars, three sizes too small. “I look like a waiter at a haunted casino,” he said. “You look unforgettable.” “That’s what worries me.” A minibus stopped outside the house. Twelve children poured out, followed by a woman carrying a stack of wrapped boxes taller than her head. Molly watched her disappear through the front door. “Revised intelligence,” she said. “There are more presents.” “Revised objection. There are more children.” “They’re not security guards.” “One of them has a foam sword. I’ve seen what they do with those.” Molly waved away his concern and marched across the street. Bill followed at the cautious pace of a man approaching a dentist who charged by the scream. Before they reached the gate, Chopper bounded over, seized Molly’s dangling tablecloth, and galloped into the garden. “Behold,” Bill said weakly. The children shrieked with delight. Chopper dragged the cloth around the lawn while Molly chased him, scattering cards from her sleeves. Parents applauded from the patio. Alfie’s mother hurried over. “Are you the magician?” Molly retrieved her hat from a flowerbed and smiled. “Of course.” “Wonderful. Alfie’s been waiting all morning.” Bill glanced through the open door at the mountain of presents. Then Alfie handed him a biscuit shaped like a star. “Can you make my dog disappear?” Alfie asked. Bill considered Chopper, Molly, and the biscuit. “No,” he said. “But I can make this disappear.” The children cheered as he ate it. Molly bowed beside him, already planning a second trick.
created_at
"2026-09-23T16:34:42.329Z"
9/23/2026, 4:34:42 PM
model
"openai/gpt-6-sol-20260922"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2217
moderation_latency
(null)
generation_time
15797
tokens_prompt
75
tokens_completion
879
native_tokens_prompt
66
native_tokens_completion
708
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
0.007212
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790181282-c2EZS9LGDfGT465dItMZ"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790181282-SKgjK3nJuF2Ptr0yp3JN"
upstream_id
"resp_07e86f3869b022bd016ab3ffa2865487d1b579dca1d2856283"
provider_responses
0
endpoint_id
"1cfc7d9d-4404-4b8e-9ee4-58ae45c9dcd4"
id
"resp_07e86f3869b022bd016ab3ffa2865487d1b579dca1d2856283"
is_byok
false
latency
565
model_permaslug
"openai/gpt-6-sol-20260922"
provider_name
"OpenAI"
status
200
total_cost
0.007212
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
20.19%
Matches word count
n/a
neededClean
false
words
520
2.00%
Dialogue to Total Word Ratio
Ratio: 41.15%, Deviation: 8.85%
neededClean
false
wordsTotal
520
wordsDialogue
214
11.0941%