NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-09-01KX415TQ08NF10ZAVGZ27200F
Run Details
Model:
GPT-5.6 Sol (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
1m 19s
Cost:
$0.176790
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
65.51%
Dialogue to Total Word Ratio
(details)
82.7557%
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 and Bill crouched behind a hedge opposite a garden exploding with balloons, cake, and children. Between them lay two sacks, three disguises, and a purple-crayon floor plan. “That is not a blueprint,” Bill said. “That is a pirate map.” “The birthday girl drew it for me.” “You spoke to the target?” “I asked where they keep presents.” “And she answered?” “She charged me two jelly beans. Ruthless child. I respect her.” “Explain the plan.” “We enter through the gate.” “That is how guests enter.” “Exactly. Nobody suspects people using the intended entrance.” “We’re wearing masks.” “Party entertainers wear masks.” “Which entertainers?” “Educational raccoons.” “We have no raccoon costumes.” Molly lifted two striped oven mitts. “Behold: tails.” “Those are oven mitts.” “Raccoons are basically burglars with oven mitts. The resemblance will create trust.” Bill peered through leaves. A magician entertained the children beside a mountain of wrapped boxes. One grandmother guarded the buffet with military stillness. “Security?” “Magician, twelve parents, grandmother at nine o’clock.” “Threat assessment?” “Magician has concealment skills. Parents are distracted. Grandmother frightens me.” “Same. We’ll avoid the sausage rolls.” “I wasn’t looking at them.” “You’ve been looking at them for four minutes.” “I’m surveilling the mustard.” “The mustard is not wearing a cardigan and glaring at us.” “Perhaps it should be. Condiments lack vigilance.” “The presents are under the banner. We cross the lawn, fill the sacks, and depart before cake.” “Why before cake?” “Children become faster after icing.” “Scientific?” “I lost a shoe at a christening.” “That explains nothing.” “It explains why I own one brown loafer.” “What are our aliases?” “Mr. and Mrs. Snickerdoodle.” “We are not married.” “Good. Let resentment season the performance.” “I want my own surname.” “Choose quickly.” “Thunder.” “Bill Thunder?” “Doctor Bill Thunder.” “You’re playing a raccoon.” “A raccoon can earn a doctorate.” “Not in oven mitts.” “That prejudice is why woodland education remains underfunded.” A fairy-winged girl wandered toward the hedge. Molly smiled weakly. Bill hid the sacks behind his back, making himself twice as suspicious. The girl stared at the oven mitts, accepted this as adult behavior, and skipped away. Bill exhaled into the hedge. “Abort.” “She didn’t recognize us.” “She recognized something and chose mercy.” “Stick to the plan.” “What if someone asks which child is ours?” “Point vaguely and say, ‘The sticky one.’” “They’re all sticky.” “Exactly. Universal parenthood.” “And if the magician notices us?” “Applaud. Magicians cannot distinguish admiration from surrender.” “And Grandmother?” “Offer her Bill Thunder.” “Bill Thunder refuses.” “Doctor Bill Thunder understands that fieldwork carries risks.” “Doctor Bill Thunder is unionized.” “You’re self-employed.” “Then I’m striking.” “You can strike after we steal the presents.” “That sentence contains a troubling management philosophy.” “You’ll receive half the dolls and first choice of puzzles.” “Dental?” “Jigsaw.” “Deal.” Molly tucked the oven mitt into Bill’s belt and shoved him through the gate. A father immediately handed him a plate of sausages. Bill looked back, triumphant. “See?” Molly whispered. “Perfect disguise.” Bill raised the plate. “I’m immediately retiring.”
created_at
"2026-07-09T18:10:27.177Z"
7/9/2026, 6:10:27 PM
model
"openai/gpt-5.6-sol-20260709"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
4114
moderation_latency
(null)
generation_time
79095
tokens_prompt
77
tokens_completion
1484
native_tokens_prompt
66
native_tokens_completion
5882
native_tokens_completion_images
0
native_tokens_reasoning
5146
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
0.17679
router
(null)
provider_responses
0
endpoint_id
"a54c5de0-89bf-4ad7-a212-cf977eed918a"
id
"resp_08ce5ccb7e4a7cda016a4fe41344a0819398401da96d9618da"
is_byok
false
latency
238
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-1783620627-sPSectfQIUOhgLt7mBuC"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1783620627-WKwpGPPcdSpm4JWJ4MrO"
upstream_id
"resp_08ce5ccb7e4a7cda016a4fe41344a0819398401da96d9618da"
total_cost
0.17679
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
100.00%
Matches word count
n/a
neededClean
false
words
499
65.51%
Dialogue to Total Word Ratio
Ratio: 72.91%, Deviation: 2.91%
neededClean
false
wordsTotal
502
wordsDialogue
366
82.7557%