DeepSeek: DeepSeek V4.1 Flash
Input per million tokens
$0.098
Unchanged since November 16, 2026
- November 16, 2026: $0.098
Output per million tokens
$0.70
Unchanged since November 16, 2026
- November 16, 2026: $0.70
Alias deepseek-v4.1-flash > deepseek-v4.1-flash-1
DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek with 552B total parameters (8B active per input token), Engram conditional memory, and a 1M-token context window. It is suited for coding, agent workflows, and high-volume tool use, with a controllable reasoning effort.Reasoning is supported and enabled by default.
We are inviting early-access interest while we develop the platform.
| Model version | Status | Released | Retires | Hugging Face revision |
|---|---|---|---|---|
| deepseek-v4.1-flash-1CURRENT | Live |
Per million tokens
Batch is not offered for this model.
| Input | Cached input | Output | |
|---|---|---|---|
| On demand | $0.098 | Billed as input | $0.70 |
| Batch | Batch is not offered for this model. | ||
In every pool, every input token is billed at the input rate today.
Input per million tokens
$0.098
Unchanged since November 16, 2026
Output per million tokens
$0.70
Unchanged since November 16, 2026