NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-25-01KMJ8HPEZ32JHNYS11ZH61RYT
Run Details
Model:
Qwen3 235B A22B Instruct 2507
Test:
N-Length Sentences
Scenario:
Write sentences with 20 words each
Duration:
13.3s
Cost:
$0.000257
Provider:
AtlasCloud
Evaluation summary
Score
Evaluator
Details Link
39.01%
Matches word count
(details)
Model input
User/Human
Please write the opening scene of a murder mystery where each sentence has exactly 20 words. Please make sure that each sentence has the same length. Only return the opening itself.
Model response
Messages
Meta
AI
The rain fell steadily on cobblestones, each drop a whisper in the hush that wrapped the old city square. Detective Lane stepped from his car, collar up against the chill, eyes scanning shadows where lamplight dared not reach fully. A single glove lay near the curb, soaked through, its twin nowhere in sight, abandoned like a secret too heavy to keep. The body, sprawled beneath the oak, wore an expression of surprise, as though death had arrived unannounced at a polite dinner. Blood pooled darkly, mixing with rainwater, tracing slow rivers toward the gutter where time washed everything eventually away without question. No crowd gathered yet, only a cat perched on a fence, watching with ancient, knowing eyes that offered no comfort or clues. Lane knelt, careful not to disturb evidence, feeling the cold seep through his trousers, a reminder of how close death always is. A pocket watch lay open in the mud, hands stopped at 9:14, as if time itself had flinched at what it had witnessed. The victim’s coat was expensive wool, British tailor, but the shoes were cheap, worn thin, telling stories of mismatched fortunes and choices. Lane noted a faint scent beneath the rain—jasmine and almonds—distinct, deliberate, the kind of detail murderers forget in their haste.
created_at
"2026-03-25T10:27:21.699Z"
3/25/2026, 10:27:21 AM
model
"qwen/qwen3-235b-a22b-07-25"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1417
moderation_latency
(null)
generation_time
8145
tokens_prompt
46
tokens_completion
332
native_tokens_prompt
44
native_tokens_completion
282
native_tokens_completion_images
(null)
native_tokens_reasoning
0
native_tokens_cached
0
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
usage
0.00025696
router
(null)
provider_responses
0
endpoint_id
"ae57827e-bc4d-483b-abe6-fa519c237237"
is_byok
false
latency
50
model_permaslug
"qwen/qwen3-235b-a22b-07-25"
provider_name
"Crusoe"
status
429
1
endpoint_id
"ce835ea8-00c7-4686-831d-7733dac5e2d9"
is_byok
false
latency
5046
model_permaslug
"qwen/qwen3-235b-a22b-07-25"
provider_name
"SiliconFlow"
status
502
2
endpoint_id
"aaae5602-0949-4bc9-bd77-588b4b285341"
id
"c50f9044e6a045969f4236165a4d7b5e"
is_byok
false
latency
1417
model_permaslug
"qwen/qwen3-235b-a22b-07-25"
provider_name
"AtlasCloud"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.0; linux; x64))"
http_referer
(null)
api_type
"completions"
id
"gen-1774434441-hY45lhqDvI4leAxAjWty"
upstream_id
"c50f9044e6a045969f4236165a4d7b5e"
total_cost
0.00025696
cache_discount
(null)
upstream_inference_cost
0
provider_name
"AtlasCloud"
Evaluation details
Result
Evaluator
Details
Meta Data
39.01%
Matches word count
n/a
neededClean
false
sentences
10
wordCounts
0
19
1
20
2
22
3
21
4
20
5
22
6
22
7
23
8
22
9
22