NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-02-01M1HAW5JK1RGWA7P3G8QW9HKB
Run Details
Model:
Qwen 3.7 Flash (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
1m 28s
Cost:
$0.001636
Provider:
Alibaba
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0000%
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
Rain drummed against the minivan roof as Bill adjusted his goggles. Molly sat beside him, arranging laminated floor plans on her lap. Beyond the fence, colorful balloons bobbed in the evening breeze above a lawn. Laughter drifted through open doors, mixing with bass speakers inside. Neon streamers swayed violently in gusts. Glitter coated the dashboard messily. Cake crumbs littered passenger mats. We initiate phase one when the host slices the cake, Bill announced, tapping his notebook. Every child watches dessert simultaneously. We gain exactly eighty seconds to clear the sector before parents notice anomalies. Eighty seconds assumes zero hesitation, she replied, rolling her shoulders against the jacket. Pressure sensitive rugs trigger alarms at three pounds. We must distribute weight evenly across rubber mats. Why did you park beneath the streetlamp? Light draws attention away from the service entrance. When crowds gather near the porch, we slip through the ivy trellis. I mapped the patrol route. The automated vacuum crosses the hallway every sixty seconds. We utilize the suction cycle to mask footsteps. Robots lack creativity. Human subjects panic when toys disappear. Children wail loudly, adults argue fiercely, and we escape before cleanup begins. Hand me the extraction bags. She tossed the canvas sacks across the console. They landed heavily against his knee. He caught them instinctively and tested the seams carefully. The fabric felt surprisingly durable. These contain precisely twenty units each, Molly observed, checking her watch. We prioritize high value electronics near the fireplace. Leaving wooden blocks behind ensures minimal suspicion. Bill secured the strap and nodded toward the garage. The hatch remained unsealed despite manufacturer specifications. He produced a tension wrench and worked the lock mechanism quietly. Metal clicked against metal inside the frame. Speed matters more than perfection, he whispered, sliding the bolt upward. Once automatic locks disengage, we flood the room and grab everything nearby. No second thoughts. Clean execution. Molly grabbed her gloves and stood. Her knees bumped against the dashboard as she prepared to step outside. The scent of wet grass and frosting filled the cabin. Recall the contingency protocol, she reminded him, adjusting her hood. If generators fail, we abandon the mission instantly. Capture risk always overrides inventory concerns. Bill kicked the door open and hopped onto the pavement. Water splashed against his boots as he surveyed the path. Thunder rumbled in clouds above rooftops. He signaled toward the alleyway. Protocol noted, he shouted over the wind. Meet me at the extraction point after desserts conclude. Do not delay for souvenirs. Molly followed him into the rain. Cold droplets soaked her hair as she sprinted toward the fence. Beyond the property, a bell chimed and cheers erupted. Mission started. Rain continued falling steadily. Moonlight illuminated puddles brightly. Winds picked up speed suddenly.
created_at
"2026-09-02T15:12:51.294Z"
9/2/2026, 3:12:51 PM
model
"qwen/qwen3.7-flash-20260727"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
399
moderation_latency
(null)
generation_time
87439
tokens_prompt
75
tokens_completion
12910
native_tokens_prompt
76
native_tokens_completion
12566
native_tokens_completion_images
(null)
native_tokens_reasoning
11996
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"
service_tier
(null)
usage
0.00163586
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788361971-6dx8BcsZRwTR6h6abroF"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788361971-FAVfvAXh00OeioMCBxfM"
upstream_id
"chatcmpl-8bb599f0-0c48-939a-b7cc-9ef37f70d499"
provider_responses
0
endpoint_id
"5a9a0ee6-874a-4d03-92be-e5094c0df4e7"
id
"chatcmpl-8bb599f0-0c48-939a-b7cc-9ef37f70d499"
is_byok
false
latency
399
model_permaslug
"qwen/qwen3.7-flash-20260727"
provider_name
"Alibaba"
status
200
total_cost
0.00163586
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Alibaba"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
459
0.00%
Dialogue to Total Word Ratio
Ratio: 0.00%, Deviation: 50.00%
neededClean
false
wordsTotal
459
wordsDialogue
0
0.0000%