NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-27-01KQ72M8QQTX4AAW0QQ0TFZ2Y9
Run Details
Model:
Qwen 3.6 35B
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
1m 14s
Cost:
$0.008598
Provider:
AtlasCloud
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
65.34%
Dialogue to Total Word Ratio
(details)
82.6695%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 10% 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 adjusted her tactical tutu, the tulle scratching against the night-vision goggles perched precariously on her forehead. Bill fumbled with the grappling hook, which was actually a modified hair dryer suction-cupped to a plastic dragon. They crouched behind a hedge shaped like a duck, watching the glow-in-the-dark bouncy castle inflate with menacing hums across the lawn. The target: the Mountain of Stuff under the tree draped in garland that smelled faintly of glitter and poor decisions. Molly checked her watch, a device that played "Baby Shark" every minute on the hour. She pointed a gloved finger at the open window, where a golden retriever was currently asleep on a plate of cookies. Bill nodded, sweating profusely through his "Super Spy" mask. Bill whispered, Did you bring the velvet sacks? Molly hissed, Stop whispering, you're scaring the dragon-dryer! Go! I found the Wii! Run! He signaled the distraction. Molly kicked a pebble toward the dog. The beast's ear twitched, but it didn't move. The dragon-dryer sputtered, launching a confetti blast that hit the retriever. Molly face-palmed. Bill scrambled into the bouncy castle. The toddler was awake. They needed the loot. The dog barked. Chaos ensued. They grabbed the gifts. Hurry! Quickly.
created_at
"2026-04-27T08:59:21.212Z"
4/27/2026, 8:59:21 AM
model
"qwen/qwen3.6-35b-a3b-20260415"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
870
moderation_latency
(null)
generation_time
73637
tokens_prompt
77
tokens_completion
7474
native_tokens_prompt
76
native_tokens_completion
8895
native_tokens_completion_images
(null)
native_tokens_reasoning
8600
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
"stop"
usage
0.008598149
router
(null)
provider_responses
0
endpoint_id
"179c6985-3a1c-48f1-a009-9f3e3e402cdd"
id
"620db5a85c26408c87a60facafb0d2d5"
is_byok
false
latency
870
model_permaslug
"qwen/qwen3.6-35b-a3b-20260415"
provider_name
"AtlasCloud"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.1; linux; x64))"
http_referer
(null)
request_id
"req-1777280361-jMWFCWzB3vdH2msSlkxq"
session_id
(null)
api_type
"completions"
id
"gen-1777280361-Xnf8EKlp4wa7UU2RYyQr"
upstream_id
"620db5a85c26408c87a60facafb0d2d5"
total_cost
0.008598149
cache_discount
(null)
upstream_inference_cost
0
provider_name
"AtlasCloud"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
200
65.34%
Dialogue to Total Word Ratio
Ratio: 1.92%, Deviation: 8.08%
neededClean
false
wordsTotal
208
wordsDialogue
4
82.6695%