NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-24-01KQ0B28R3B3SCC3ZE278TWZS4
Run Details
Model:
GPT-OSS 120B
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
2m 4s
Cost:
$0.002160
Provider:
Io Net
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
99.80%
Dialogue to Total Word Ratio
(details)
99.9004%
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 crouched behind the striped tent, her black leggings stark against pastel balloons. Sunlight flickered through the plastic canopy like a nervous disco ball, and toddlers’ giggles rose like sugary fog. Bill hovered by the punch table, his oversized cap hiding a flashlight and a half‑eaten donut. He whispered, “We have ten minutes before the clown shows up, so act fast.” Molly smirked, pulling a crumpled blueprint from her pocket; it was a doodle of the party layout with a red‑X marking the gift table. The room reeked of frosting and cheap perfume, making Bill’s stomach rumble louder than the marching band music. He checked his watch; seconds ticked down like mischievous ants. Molly adjusted the strap on her satchel, already stuffed with a few glittering trinkets from past jobs. She glanced at the inflatable castle, its bright orange arches swaying, and imagined the chaos when the presents vanished. Behind the clown’s red nose, a nervous mother chatted with a PTA volunteer about cupcakes, oblivious to the two silhouettes plotting their swoop. The party’s centerpiece was a towering dinosaur cake, frosting dripping like lava. Bill nudged Molly, “Ready?” she answered, “As ready as a kid on a sugar high.” The thieves exchanged a conspiratorial grin, eyes glittering with promise. They slipped into the shadows, footsteps muffled by a carpet of confetti, and prepared to execute the most childish heist ever imagined. Moonlight glimmered on tangled streamers, casting playful shadows that danced like mischievous pixies while the night hummed softly gently. Bill: "First we disable the clown's balloon arch, then we slip under the cake." Molly: "Got it. I’ll jam the music so the kids laugh longer." Bill: "The security guard is disguised as a dad, but his badge is fake." Molly: "We’ll distract him with a balloon animal, then swap the gift tags." Bill: "What's the plan for the presents?" Molly: "We grab the big box on the left, dump the toys into the trash, and take the candy." Bill: "And the chocolate eggs?" Molly: "We hide them in the clown’s oversized shoes." Bill: "Perfect. I’ll use the suction cup to open the gift box without breaking anything." Molly: "Remember, the kids will be singing ‘Happy Birthday’ at exactly 3:15, that’s our window." Bill: "Got the stopwatch ready?" Molly: "Timed. 30 seconds max." Bill: "Okay, on my signal, we crawl under the table and grab the ribbon-wrapped loot." Molly: "Signal?" Bill: "The clown’s squeaky shoe." Molly: "Got it. When he steps, we’ll slide like ninjas." Bill: "One… two… three… now!" Molly: "We’re in! The gift table is right there, covered by streamers." Bill: "Pull the red box, it’s the one with the golden bow." Molly: "Got it, swapping now." Bill: "Hurry, the clown’s theme song is ending." Molly: "All loot secured, moving to exit." Bill: "Great, now the fireworks!" Molly: "Wait, the police are on the speaker system." Bill: "Quick! Deploy the confetti cannon!" Molly: "Now!" Bill: "We disappear into the balloons, leaving only giggles behind." Molly: "Best. Heist. Ever. Seriously."
created_at
"2026-04-24T18:12:07.566Z"
4/24/2026, 6:12:07 PM
model
"openai/gpt-oss-120b"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
5466
moderation_latency
(null)
generation_time
123333
tokens_prompt
77
tokens_completion
6172
native_tokens_prompt
129
native_tokens_completion
6112
native_tokens_completion_images
(null)
native_tokens_reasoning
5362
native_tokens_cached
64
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
"stop"
usage
0.002160445
router
(null)
provider_responses
0
endpoint_id
"aa7799a4-4935-4de1-9cb3-8ebf9c1cf4e3"
id
"chatcmpl-104887e0-8416-4c2a-a5e3-97ecac279642"
is_byok
false
latency
5466
model_permaslug
"openai/gpt-oss-120b"
provider_name
"Io Net"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.1; linux; x64))"
http_referer
(null)
request_id
"req-1777054327-mZZfcHQH8nSJMPjcEobL"
session_id
(null)
api_type
"completions"
id
"gen-1777054327-1tkXyeEkYFdo0x4kmFZx"
upstream_id
"chatcmpl-104887e0-8416-4c2a-a5e3-97ecac279642"
total_cost
0.002160445
cache_discount
0.00000576
upstream_inference_cost
0
provider_name
"Io Net"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
501
99.80%
Dialogue to Total Word Ratio
Ratio: 50.20%, Deviation: 0.20%
neededClean
false
wordsTotal
502
wordsDialogue
252
99.9004%