DeepSeek V4.1 Flash API price, and what each result really costs
DeepSeek, listed 17 Oct 2026. Price checked 12 Nov 2026; unchanged since we started tracking it. Source: DeepSeek's pricing page. deepseek/deepseek-flash
List price per million tokens
- Input
- $0.30
- Cached input
- $0.006
- Output
- $1.20
- Context window
- 1.05M tokens
Note: Peak-hour price (Mon-Fri 01:00-04:00 and 06:00-10:00 UTC). Off-peak is half: $0.15 in / $0.003 cached / $0.60 out. API model name deepseek-flash = DeepSeek-V4.1-Flash-1017. OpenRouter headline $0.15/$0.60 matches the off-peak rate.
What it costs per task, and per task that works
Six common task shapes with typical token counts. The real cost counts the tries that get redone; the rates are examples, not measurements.
| Task | Per 1,000 tries | Per 1,000 that work, 8 in 10 | Per 1,000 that work, 6 in 10 | Per 1,000 that work, 4 in 10 |
|---|---|---|---|---|
| Support bot answer3,000 in / 300 out | $1.26 | $1.58 | $2.10 | $3.15 |
| Invoice field extraction2,500 in / 250 out | $1.05 | $1.31 | $1.75 | $2.63 |
| Lead scoring1,200 in / 100 out | $0.48 | $0.60 | $0.80 | $1.20 |
| Email reply draft1,500 in / 400 out | $0.93 | $1.16 | $1.55 | $2.33 |
| Translation900 in / 800 out | $1.23 | $1.54 | $2.05 | $3.07 |
| Search re-ranking4,000 in / 50 out | $1.26 | $1.58 | $2.10 | $3.15 |
Work out your own real cost per resultTest DeepSeek V4.1 Flash on your own prompt
Compare DeepSeek V4.1 Flash with
- DeepSeek V4 Pro$1.32 in / $3.96 outSwitch cost
- MiniMax M3$0.30 in / $1.20 outSwitch cost
- Llama 4 Maverick$0.35 in / $1.00 outSwitch cost
- Qwen3.7 Plus$0.32 in / $1.28 outSwitch cost
- Gemini 3.1 Flash-Lite$0.25 in / $1.50 outSwitch cost
Worked-through switches: GPT-4o → DeepSeek V4.1 Flash: cost and quality on real tasks, Claude Sonnet 5.5 → DeepSeek V4.1 Flash: cost and quality on real tasks
Questions
How much does the DeepSeek V4.1 Flash API cost?
$0.30 per million input tokens and $1.20 per million output tokens, or $0.006 per million for cached input. Price checked 12 Nov 2026; unchanged since we started tracking it.
What does a result that works really cost with DeepSeek V4.1 Flash?
A support bot answer (about 3,000 tokens in, 300 out) costs $0.0013 per try. If 6 in 10 work first time, each answer that works really costs $0.0021, because the redone ones were paid for too.