NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-02-01M1HC3MY2KFXBSZEXJ6DQ7MFT
Run Details
Model:
Qwen 3.7 Flash (Reasoning)
Test:
Dialogue tags
Scenario:
Write unattributed dialogue
Duration:
28.7s
Cost:
$0.000489
Provider:
Alibaba
Evaluation summary
Score
Evaluator
Details Link
100.00%
Count dialogue tags
(details)
Model input
User/Human
You are an expert novel writer in the cozy mystery genre. Write a scene of about 200 words, with dialogue between two characters. The dialogue should be unattributed, meaning the speaker is not explicitly mentioned. The scene should be set in a small town bakery. Only return the story, nothing else.
Model response
Messages
Meta
AI
Flour drifted off the cooling rack onto the checkerboard tiles, settling like snow around the boots of a nervous visitor standing too close to the oven doors. The air thick with the scent of vanilla and anxiety hung heavy. "I didn't take the locket from Mrs. Higgins' purse. You know how much I despise jewelry. Gold ruins my manicure." A sudden gust from the propped-open door whipped through the shop, curling the corner of a parchment slip wedged deep within the jar of snickerdoodles. Calloused fingers extended, prying the note loose before brushing crumbs from its surface with a tea towel. "Liar. That note confesses everything, and the ink matches the smudge on your cufflink." The visitor stumbled back, leather soles scraping harsh against the floorboards, crashing hip-first into the pastry case. Crystal glasses shivered; a tower of éclairs teetered precariously. "Anyone could forge that script!" The voice rose, shrill and desperate, fighting the low thrum of the proofing baskets. "As for the cufflinks, they're family heirlooms! You're reaching here. Just admit you framed me so you could keep the deed to the mill!" Tea cloth balled tight in a fist, knuckles white against the red flannel of an apron. A crooked grin revealed teeth stained with beet sugar. "The deed isn't worth much anymore, especially once the county discovers the arsenic traces in the almond extract batch stored under your stallion. Help me turn yourself in before the sheriff tastes the latest batch of macaroons, or the whole town will know you baked the murder weapon."
created_at
"2026-09-02T15:34:24.979Z"
9/2/2026, 3:34:24 PM
model
"qwen/qwen3.7-flash-20260727"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
441
moderation_latency
(null)
generation_time
28653
tokens_prompt
78
tokens_completion
4079
native_tokens_prompt
78
native_tokens_completion
3747
native_tokens_completion_images
(null)
native_tokens_reasoning
3396
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
0.00048945
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788363264-euWFlkiQeJ5CGjZ8Q4ce"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788363264-oIpKMsdYeMVLGbGPNYpw"
upstream_id
"chatcmpl-1a3b348c-1559-96cf-b2fc-de1a81a17cd5"
provider_responses
0
endpoint_id
"5a9a0ee6-874a-4d03-92be-e5094c0df4e7"
id
"chatcmpl-1a3b348c-1559-96cf-b2fc-de1a81a17cd5"
is_byok
false
latency
438
model_permaslug
"qwen/qwen3.7-flash-20260727"
provider_name
"Alibaba"
status
200
total_cost
0.00048945
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
100.00%
Count dialogue tags
n/a
neededClean
false
dialogueTags
(empty)