NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHYB2RXWW2VG58GPWC61K9A3
Run Details
Model:
Ministral 3 14B
Test:
Dialogue tags
Scenario:
Write unattributed dialogue
Duration:
2.9s
Cost:
$0.000077
Provider:
Mistral
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
The bell above the door jingled as the morning chill slipped inside, carrying the scent of cinnamon and buttery pastry. A loaf of sourdough sat between them on the counter, its crust golden and cracked just enough to tempt. "You always bring me the good stuff." "Wouldn’t dream of bringing you anything less. Especially not after what happened last time." A chuckle, low and warm. "Ah, yes. The time you swore those blueberry muffins were cursed." "Not cursed. *Haunted.* By my own poor baking choices." The other leaned in, voice dropping to a conspiratorial whisper. "Between us, I think the real curse was that detective’s questions. You were sweating through your apron." "Detective? Please. I was just—hot. From the oven." A pause. The clink of a coffee cup being set down. "So. You still haven’t figured out who took the town’s prize-winning honey jar?" "Would I be asking you for a second cup of coffee if I had?" The laughter bubbled up, rich and unguarded. "Fair point. But next time, maybe don’t leave the evidence in the display case with a *note*." "In my defense, it was a *very* polite note." "Polite? It said *‘Thanks for the honey, you’re all sticky.’*" Silence. Then, from behind the counter, the sound of a muffled groan. "I *hate* puns."
created_at
"2026-02-20T20:15:26.148Z"
2/20/2026, 8:15:26 PM
model
"mistralai/ministral-14b-2512"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
228
moderation_latency
(null)
generation_time
2753
tokens_prompt
81
tokens_completion
342
native_tokens_prompt
68
native_tokens_completion
318
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
0.0000772
router
(null)
provider_responses
0
endpoint_id
"0b53717b-9af5-4de3-80af-ebedd2adf981"
id
"41ea4b8f0c7d41a193acd1ba364f418b"
is_byok
false
latency
228
model_permaslug
"mistralai/ministral-14b-2512"
provider_name
"Mistral"
status
200
api_type
"completions"
id
"gen-1771618526-mLPTzYWinLZLF2E40Bvf"
upstream_id
"41ea4b8f0c7d41a193acd1ba364f418b"
total_cost
0.0000772
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Mistral"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Count dialogue tags
n/a
neededClean
false
dialogueTags
(empty)