NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-27-01KQ73QT5MNRQNNKNECZZ7070T
Run Details
Model:
Qwen 3.6 27B
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
2m 31s
Cost:
$0.040747
Provider:
Alibaba
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
99.19%
Dialogue to Total Word Ratio
(details)
99.5966%
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
The blueprints of Henderson Manor lay across the cracked diner table. Molly traced an oily finger along the perimeter fence, noting the poorly secured side gate. Bill adjusted his oversized fake mustache, nearly choking on lukewarm coffee while studying the supply manifest. Three hundred wrapped boxes awaited inside the ballroom, stacked like a fortress of foil. Security consisted mainly of a yapping poodle and exhausted parents sipping wine. Molly adjusted her tactical visor, checking thermal scopes for nonexistent motion sensors. Bill calibrated the magnetic grapple line, muttering about load limits and glitter glue. The operation required absolute silence, perfect timing, and zero regard for childhood joy. Molly packed insulated duffel bags while Bill tested smoke pellets. She ensured they would clear the room without triggering alarms. The heavy wall clock ticked toward midnight, when the ceremony would begin. Molly tightened her harness, ready to swing across the manicured lawn. Bill carefully loaded specialized tranquilizers for large dogs. The cold pale moonlight illuminated their ridiculous equipment and shared grim determination. They prepared to breach the perfectly manicured suburban paradise. They waited. "Secure the rope ladder first," Molly said. "I will disable the cameras," Bill replied. "Grab the glitter bows," she added.
created_at
"2026-04-27T09:18:45.946Z"
4/27/2026, 9:18:45 AM
model
"qwen/qwen3.6-27b-20260422"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
681
moderation_latency
(null)
generation_time
150653
tokens_prompt
77
tokens_completion
8987
native_tokens_prompt
76
native_tokens_completion
11306
native_tokens_completion_images
(null)
native_tokens_reasoning
11049
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
0.0407472
router
(null)
provider_responses
0
endpoint_id
"82097366-3a73-4790-ac36-f96ba9f6f577"
id
"chatcmpl-2bd0aef9-a077-94b2-8b47-811a12368b8b"
is_byok
false
latency
681
model_permaslug
"qwen/qwen3.6-27b-20260422"
provider_name
"Alibaba"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.1; linux; x64))"
http_referer
(null)
request_id
"req-1777281525-6stgWf4USHZmReoclxNc"
session_id
(null)
api_type
"completions"
id
"gen-1777281525-YprnBzWxcmdV2h6BVJtH"
upstream_id
"chatcmpl-2bd0aef9-a077-94b2-8b47-811a12368b8b"
total_cost
0.0407472
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Alibaba"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
200
99.19%
Dialogue to Total Word Ratio
Ratio: 7.00%, Deviation: 3.00%
neededClean
false
wordsTotal
200
wordsDialogue
14
99.5966%