deepseek-v4-flash-0731· Released Jul 31, 2026
deepseek/deepseek-v4-flash-0731
DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total. This re-post-trained revision is suited for coding, reasoning, and agent workflows.
Input
From$0.154/M
Output
From$0.308/M
Context
Up to 1M
Max output
Up to 384K
Modalities
TextTextInput and output show the lowest price across upstreams, for reference. Your actual prices are listed per upstream below.
Upstreams & pricing
| Upstream | Context | Max output | Input $/1M | Output $/1M | Cached input $/1M |
|---|---|---|---|---|---|
ScitiXFrom price | 1M | 384K | $0.154 | $0.308 | $0.033 |
Paratera (CN) | 1M | 384K | $0.476141 | $1.42841 | $0.015876 |
Routed in real time — you pay the listed price of whichever upstream serves your call.
How to call
curlrequest.sh
curl https://api.tkex.ai/v1/chat/completions \
-H "Authorization: Bearer $TX_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek/deepseek-v4-flash-0731",
"messages": [
{"role": "user", "content": "Say hello in five words."}
]
}'