NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-09-01KX413GVM0W4EJ494QHVT192J
Run Details
Model:
GPT-5.6 Luna
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
7.5s
Cost:
$0.004404
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
0.03%
Matches word count
(details)
93.02%
Dialogue to Total Word Ratio
(details)
46.5249%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 30% 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
Rain drummed on the van roof while Molly unfolded a blueprint across the dashboard. It was actually a takeaway menu, but the grease stain resembled the party hall well enough. At midnight, the community center glowed beneath a banner reading HAPPY BIRTHDAY, TREVOR! Balloons bobbed behind the windows, and a cardboard dragon guarded the entrance with an expression of permanent disappointment. Bill adjusted his tiny black mask. “Are we sure this is a heist?” “It has a target, security, and a getaway vehicle,” Molly said. “That is the textbook definition.” “Our getaway vehicle is a borrowed ice-cream van.” “Exactly. Nobody suspects ice cream.” “They suspect ice cream when it’s raining.” Molly pointed to the side door. “We enter through the kitchen, cross the corridor, and reach the presents before the magician’s second act.” Bill studied the map. “What’s the magician’s first act?” “Making a rabbit disappear.” “And the second?” “Making a balloon poodle.” “That doesn’t sound dangerous.” “Neither does a rabbit until you’re hiding from it in a cupboard.” They slipped from the van wearing matching yellow raincoats and carrying a crate labeled EMERGENCY PUDDING. The label was Molly’s idea. Bill had wanted “Highly Classified Cake,” but Molly said cakes attracted questions. The kitchen door yielded to a bent spoon. Inside, three bakers were arguing over a mountain of cupcakes. One glanced up. “Delivery?” he asked. “Pudding emergency,” Molly replied. The baker nodded solemnly and returned to the argument. They moved through the corridor, past a table where children had abandoned paper crowns, plastic swords, and one alarming number of raisins. At the end stood the present table, buried under brightly wrapped boxes. Bill’s eyes widened. “That’s a fortune.” “It’s mostly socks and educational toys.” “Fortunes come in many forms.” A shriek erupted from the party room. The magician had begun his second act. A balloon animal squealed as he twisted it into shape. Molly opened the emergency crate. Inside were two oven mitts, a rubber chicken, and a small hand vacuum. “Why the chicken?” Bill whispered. “Diversion.” She squeezed it. The chicken produced a noise like a goose being audited. Every adult turned toward the kitchen. Bill grabbed the nearest presents, while Molly swept smaller packages into a laundry sack. They worked with impressive speed until a little girl in a glittery crown appeared beside them. “What are you doing?” she asked. Bill froze. “We’re… present inspectors.” “Are you stealing them?” Molly leaned closer. “Temporarily relocating them for safety.” The girl considered this. “Can I come?” “No,” Molly said. “Yes,” Bill said. Molly glared at him. The girl pointed toward the window. “The real thieves are leaving through the front.” Molly and Bill spun around. Two adults were sprinting away with the largest boxes, pursued by a furious clown. Bill dropped the sack. “We’ve been robbed.” “By professionals,” Molly said. The birthday boy charged into the corridor. “Those are my presents!” Molly grabbed Bill’s arm. “New plan. We steal them back.” “Does that still make us thieves?” “Absolutely.” They raced into the rain, followed by the glitter-crowned girl, the clown, and six children wielding plastic swords. Behind them, the cardboard dragon toppled over, apparently unwilling to miss the adventure.
created_at
"2026-07-09T18:09:11.552Z"
7/9/2026, 6:09:11 PM
model
"openai/gpt-5.6-luna-20260709"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
492
moderation_latency
(null)
generation_time
7174
tokens_prompt
77
tokens_completion
979
native_tokens_prompt
66
native_tokens_completion
723
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
0.004404
router
(null)
provider_responses
0
endpoint_id
"eebea444-34a5-4642-a766-cb319471d33a"
id
"resp_0fc8099649dde1be016a4fe3c7a1948193932b87b9c2cef583"
is_byok
false
latency
235
model_permaslug
"openai/gpt-5.6-luna-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-1783620551-agPlNK8UwtqSRltGTLp7"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1783620551-Iy0kt1MFQqneziOJv7wi"
upstream_id
"resp_0fc8099649dde1be016a4fe3c7a1948193932b87b9c2cef583"
total_cost
0.004404
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
0.03%
Matches word count
n/a
neededClean
false
words
530
93.02%
Dialogue to Total Word Ratio
Ratio: 31.20%, Deviation: 1.20%
neededClean
false
wordsTotal
532
wordsDialogue
166
46.5249%