// pricing · September 2026
DeepSeek API Pricing
Every DeepSeek model's list price per 1M tokens — input, output and cached input — with context windows, LiveBench scores and what a real workload costs per month.
15 DeepSeek models are listed, with a median list price of $0.270 per 1M input tokens and $0.890 per 1M output tokens. Paid prices run from $0.061 (DeepSeek V4 Flash 0423) to $1.15 (R1) per 1M tokens blended.
- Models
- 15
- Median input / 1M
- $0.270
- paid models
- Median output / 1M
- $0.890
- paid models
- Cheapest paid
- $0.061
- DeepSeek V4 Flash 0423
- Highest LiveBench
- 81.1
- DeepSeek V4.1 Flash
DeepSeek model prices
List prices in USD per 1M tokens. “Blended” is a 3:1 input:output mix; the per-month column prices the selected workload using each model's cached-input rate where it publishes one.
Showing 15 of 15 models · per-month column: 8K in / 600 out × 100K requests, 50% of input cached
| deepseek/deepseek-v4-flash | $0.049 | $0.098 | $0.0098 | $0.061 | 1.0M | 65.5 | $29.40 |
|---|---|---|---|---|---|---|---|
| deepseek/deepseek-v4-flash-0731 | $0.040 | $0.640 | $0.016 | $0.190 | 1.3M | 74.2 | $60.80 |
| deepseek/deepseek-v4.1-flash | $0.100 | $0.500 | $0.010 | $0.200 | 1.0M | 81.1 | $74.00 |
| deepseek/deepseek-v3.2 | $0.269 | $0.400 | $0.134 | $0.302 | 164K | — | $185.40 |
| deepseek/deepseek-v3.2-exp | $0.270 | $0.410 | — | $0.305 | 164K | — | $240.60 |
| deepseek/deepseek-v4-flash-vision-exp | $0.220 | $0.660 | $0.0070 | $0.330 | 1.0M | 76.8 | $130.40 |
| deepseek/deepseek-chat-v3.1 | $0.250 | $0.950 | $0.130 | $0.425 | 164K | — | $209.00 |
| deepseek/deepseek-chat-v3-0324 | $0.250 | $1.00 | — | $0.438 | 164K | — | $260.00 |
| deepseek/deepseek-v3.1-terminus | $0.270 | $1.00 | $0.135 | $0.453 | 164K | — | $222.00 |
| deepseek/deepseek-chat | $0.320 | $0.890 | — | $0.462 | 164K | — | $309.40 |
| deepseek/deepseek-r1-distill-llama-70b | $0.800 | $0.800 | — | $0.800 | 8K | — | $688.00 |
| deepseek/deepseek-r1-0528 | $0.500 | $2.15 | $0.350 | $0.913 | 164K | — | $469.00 |
| deepseek/deepseek-v4-pro-0813 | $0.660 | $1.98 | $0.022 | $0.990 | 1.0M | 77.4 | $391.60 |
| deepseek/deepseek-v4-pro | $0.865 | $1.73 | $0.072 | $1.08 | 1.0M | 71.6 | $478.80 |
| deepseek/deepseek-r1 | $0.700 | $2.50 | — | $1.15 | 64K | — | $710.00 |
DeepSeek pricing FAQ
What is the cheapest DeepSeek model?
DeepSeek V4 Flash 0423 is the cheapest paid DeepSeek model, at $0.049 per 1M input tokens and $0.098 per 1M output tokens.
How much does the most capable DeepSeek model cost?
The highest LiveBench overall score among DeepSeek models belongs to DeepSeek V4.1 Flash (81.1), priced at $0.100 input and $0.500 output per 1M tokens. A RAG assistant at 100K requests a month would list at about $74.00.
Why do output tokens cost more than input tokens?
Generating a token takes a full forward pass per token, while input is processed in parallel. Across the paid DeepSeek models listed here the median output price is $0.890 per 1M against $0.270 for input — reasoning tokens bill as output too.
Which DeepSeek models support prompt caching?
10 of 15 publish a separate cached-input price, which bills repeated prompt prefixes at a fraction of the normal input rate. Filter the table by "Caching" to see them.
How current are these prices?
List prices are pulled from OpenRouter’s public models API and refreshed every 15 minutes. Batch discounts and enterprise pricing are not included.
Other providers
What these prices include
- List prices from OpenRouter's public models API, which mirrors provider pricing. Batch discounts and negotiated enterprise rates are not included.
- Variants —
:freeand:batchlistings are folded into their base model; its model page lists them. - Per-month cost is a floor: tokens you describe times list price. Cache writes and reasoning-token volume depend on your workload.