NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHM5GFTVHNN26FD0M84Y12GC
Run Details
Model:
DeepSeek V3.2
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
2m 33s
Cost:
$0.003193
Provider:
SiliconFlow
Model input
User/Human
You are a author of comedic heist books. Please write a 200 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
Molly smoothed the wrinkled, candy-smeared blueprint onto the table. "Operation Pinata Pinch is a go. Tomorrow, three PM, at the Johnson residence. Presents are deposited beside the snack table." Bill, nursing a cup of cold coffee, frowned. "A children's party? Molly, we stole a van full of televisions last week." "And we almost got caught because you left the GPS on. This is subtle. We infiltrate as party planners." She produced two laminated badges. "I'm 'Miss Joy', you're 'Mr. Fun'. We coordinate games, then during the cake chaos, we box up the gifts and walk out." "Cake chaos?" Bill set his mug down. "What's the exit strategy?" "Confusion. We'll trigger a glitter bomb near the piƱata. While everyone is dazzled, we load the loot into our 'Party Supplies' van." "A glitter bomb? That's not a weapon, it's a craft store accident." "It's non-violent and effective. Now, memorize the schedule: one PM, set up. Two, greet guests. Three, cake and glitter. Four, profit." She beamed. Bill sighed. "I feel like we've hit a new low." "Nonsense! This is a high-stakes, low-security gig. Now, practice your cheerful wave." Molly demonstrated, fingers wiggling with manic enthusiasm. Bill gave a half-hearted flap of his hand. "I need a new line of work." "After this, we'll afford one. Now, again: be joyful!"
created_at
"2026-02-16T21:28:13.368471+00:00"
2/16/2026, 9:28:13 PM
model
"deepseek/deepseek-v3.2-20251201"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1477
moderation_latency
(null)
generation_time
152757
tokens_prompt
77
tokens_completion
7554
native_tokens_prompt
67
native_tokens_completion
7559
native_tokens_completion_images
(null)
native_tokens_reasoning
7234
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
0.00319287
router
(null)
provider_responses
0
status
429
is_byok
false
latency
59
endpoint_id
"f270e10b-1d21-4345-9e4e-662ddb5ba441"
provider_name
"DeepInfra"
model_permaslug
"deepseek/deepseek-v3.2-20251201"
1
id
"019c68584069c31f7ca79abecbbcabb9"
status
200
is_byok
false
latency
1473
endpoint_id
"b39a588a-b338-43ba-b046-226b9e24afa8"
provider_name
"SiliconFlow"
model_permaslug
"deepseek/deepseek-v3.2-20251201"
api_type
"completions"
id
"gen-1771277139-RdGXFzwHFqDjs5x8SsKr"
upstream_id
"019c68584069c31f7ca79abecbbcabb9"
total_cost
0.00319287
cache_discount
(null)
upstream_inference_cost
0
provider_name
"SiliconFlow"
Evaluation details
Result
Evaluator
Details
Meta Data
9.6081%
Matches word count
n/a
neededClean
false
words
222
0.0000%
Dialogue to Total Word Ratio
Ratio: 78.38%, Deviation: 28.38%
neededClean
false
wordsTotal
222
wordsDialogue
174
4.8041%