DeepSeek V4 Flash API pricing — cost per million tokens
DeepSeek's DeepSeek V4 Flash costs $0.44 per million input tokens and $1.32 per million output tokens on the standard API tier (verified 2026-09-02). Note: Peak-hour price since 2026-08-16; off-peak is half: $0.22 / $0.66. Cache-hit input $0.014. Compare it in the full pricing table or price your own prompts in the token cost calculator.
DeepSeek V4 Flash price per million tokens
| direction | $ / Mtok | $ / 1K tokens |
|---|---|---|
| input (prompt) | $0.44 | $0.00044 |
| output (completion) | $1.32 | $0.00132 |
What real requests cost on DeepSeek V4 Flash
| workload | input | output | total |
|---|---|---|---|
| Chat message (1K in / 500 out) | $0.00044 | $0.00066 | $0.0011 |
| Long document (100K in / 2K out) | $0.044 | $0.00264 | $0.0466 |
| Agent session (500K in / 50K out) | $0.22 | $0.066 | $0.29 |
How DeepSeek V4 Flash pricing compares
On the same chat-message workload, DeepSeek V4 Flash costs 1.4× more than GPT-5.6 Luna (the cheapest tracked model) — capability, latency and context limits are the other side of that trade. Output tokens dominate long generations: at $1.32/Mtok, a 50K-token agent transcript costs $0.066 in output alone. Official rate card: DeepSeek pricing.
Related: DeepSeek V4 Pro pricing · GPT-5.6 Luna pricing · GPT-5.4 nano pricing · release timeline
The weekly e/acc newsletter
One email a week: what accelerated.
Frontier releases, price drops, compute buildouts — the week's acceleration in five minutes, sourced and numeric. Free.