NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M37HAWFC3FN74DF9WT0AG9V6
Run Details
Model:
GPT-6 Sol (Reasoning, Medium)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
18.6s
Cost:
$0.009072
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
99.98%
Matches word count
(details)
0.85%
Dialogue to Total Word Ratio
(details)
50.4187%
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 the invitation across the café table, pinning one corner beneath a sugar bowl. A cartoon dragon smiled beside the words “BIRTHDAY PARTY.” She had circled the start time in red. “We go in as guests,” she said. “Children’s parties are chaos. Nobody remembers who arrived with whom.” Bill studied the dragon. “The invitation says costumes encouraged. What are we wearing?” “Something inconspicuous.” “You brought a dragon tail.” “It’s detachable.” Bill glanced at the tail poking from Molly’s handbag. The café’s ceiling fan stirred its green felt spikes. For three days, she had spoken of nothing but the mountain of presents visible through the party venue’s window. “All the presents?” he asked. “Even the small ones?” “Especially those. Big boxes are usually educational.” “What if someone asks which child we belong to?” Molly turned over the invitation. On the back, in crayon, was a drawing of six children and a dog with eight legs. She examined each face. “We point at a child who’s running,” she said. “That could be any of them.” “Exactly.” Bill leaned back as a waiter passed their table. He lowered his voice, though the only other customer was wearing headphones and asleep. Molly drew a second circle on the invitation, this one around the words “parents welcome.” “Parents,” Bill said. “There’s the problem. They notice things.” “They notice whether everyone has eaten cake. We’ll eat cake.” “I thought we were stealing presents.” “We can be versatile.” Outside, a delivery van stopped beside the party venue. Its driver carried in an enormous wrapped parcel, nearly as tall as the doorway. Bill watched it disappear inside. He had begun to hope the job might be cancelled by common sense. “How do we carry that?” he asked. Molly folded the invitation. “We don’t. It’s too obvious.” “You said all the presents.” “I was expressing ambition, not making a legally binding promise.” “That sounds unlike you.” “It’s a new approach.” A little girl appeared at the venue window and pressed a paper crown against the glass. Behind her, adults shifted tables into a circle. Bill could see a pile of gifts beneath a banner, each one wrapped in the same blue paper. “Why are they all identical?” he asked. Molly went still. She pulled out her phone and found the venue’s party announcement. Her lips moved as she read. “It’s a present exchange,” she said. “Every guest brings a gift. Every guest leaves with one.” “So if we go in as guests…” “We have to bring gifts.” Bill looked at the sugar bowl. Molly looked at the tail. Neither seemed likely to delight a child, though he suspected the sugar had better odds. “We could buy two,” Bill said. “Then attend, eat cake, and leave with two different ones.” “That isn’t stealing.” “No,” he said. “But you’d get cake.” Molly considered this carefully. Then she picked up her handbag, its tail dragging across the floor. “Fine. But I’m choosing the gifts. Last time you bought a child a desk calendar.”
created_at
"2026-09-23T16:24:44.275Z"
9/23/2026, 4:24:44 PM
model
"openai/gpt-6-sol-20260922"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
5947
moderation_latency
(null)
generation_time
18583
tokens_prompt
75
tokens_completion
1096
native_tokens_prompt
66
native_tokens_completion
894
native_tokens_completion_images
0
native_tokens_reasoning
244
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
0.009072
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790180684-8wPJgZ6iqQ4nfCP978No"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790180684-HowO8Vl1O3TFZVyr0HIV"
upstream_id
"resp_08d46e98313e9d1c016ab3fd4c6c4c87d1bb73b507c85afb41"
provider_responses
0
endpoint_id
"1cfc7d9d-4404-4b8e-9ee4-58ae45c9dcd4"
id
"resp_08d46e98313e9d1c016ab3fd4c6c4c87d1bb73b507c85afb41"
is_byok
false
latency
633
model_permaslug
"openai/gpt-6-sol-20260922"
provider_name
"OpenAI"
status
200
total_cost
0.009072
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
99.98%
Matches word count
n/a
neededClean
false
words
502
0.85%
Dialogue to Total Word Ratio
Ratio: 40.24%, Deviation: 9.76%
neededClean
false
wordsTotal
502
wordsDialogue
202
50.4187%