All models
DeepSeek · text
DeepSeek V4 Flash Latest
Newdeepseek-v4-flash-latest
Always routes to the latest DeepSeek V4 Flash release.
Pricing
Input$0.1899/M
Output$0.3798/M
Billed 1:1 against prepaid credits — no markup, no subscription.
Tokens (30d)
13.49M
processed via gateway
Throughput
134 tok/s
typical output speed
Latency
567ms
p95 907ms
Success
99.5%
2xx responses
Usage & performance
Representative figures based on recent gateway activity.
Token throughput
Last 30 daysLatency
Last 30 daysSpecifications
Model IDdeepseek-v4-flash-latest
ProviderDeepSeek
ModalityText & code
Endpoint/api/v1/chat/completions
Context window1,000,000 tokens
Price$0.1899/$0.3798 per 1M tokens
Supported parameters
messagestemperaturemax_tokensresponse_formattoolstool_choicetop_p
How to use
Call the recommended endpoint for deepseek-v4-flash-latest.
curl https://api.blackmagic.engineering/v1/chat/completions \
-H "Authorization: Bearer $BLACKMAGIC_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek-v4-flash-latest",
"messages": [
{ "role": "user", "content": "Draft a launch note." }
]
}'Use the endpoint shown in the sample for this model.