DeepSeek API pricing per million tokens
On 10 October 2026, DeepSeek models on this list cost from $0.0059 per million input tokens (DeepSeek V4 Flash 0731) to $0.70 (R1), and from $0.17 per million output tokens (DeepSeek V4 Flash 0423) to $3.48 (DeepSeek V4 Pro 0423).
| Model | Input | Output | Cached input | Batch in / out | Context | Your month | Listed |
|---|---|---|---|---|---|---|---|
| DeepSeek V4.1 Flash | $0.15 | $0.60 | $0.003 | $0.112 / $0.336 | 1.05M | $18.00 | 2026-09-10 |
| DeepSeek V4 Flash Vision Exp DeepInfra, cheapest of 3 hosts | $0.44 | $1.32 | $0.014 | 1.05M | $46.20 | 2026-08-21 | |
| DeepSeek V4 Pro 0813 | $0.66 | $1.98 | $0.022 | 1.05M | $69.30 | 2026-08-12 | |
| DeepSeek V4 Flash 0731 OpenInference, cheapest of 24 hosts | $0.0059 | $0.27 | $0.0043 | 1.05M | $4.41 | 2026-07-31 | |
| DeepSeek V4 Pro 0423 Parasail, cheapest of 15 hosts | $0.45 | $3.48 | $0.10 | 1.05M | $79.20 | 2026-04-24 | |
| DeepSeek V4 Flash 0423 Wafer, cheapest of 16 hosts | $0.035 | $0.17 | $0.013 | 1.05M | $4.65 | 2026-04-24 | |
| DeepSeek V3.2 DeepInfra, cheapest of 12 hosts | $0.26 | $0.38 | $0.13 | 164K | $21.30 | 2025-12-01 | |
| DeepSeek V3.2 Exp served by SiliconFlow | $0.27 | $0.41 | 164K | $22.35 | 2025-09-29 | ||
| DeepSeek V3.1 Terminus SiliconFlow, cheapest of 2 hosts | $0.27 | $1.00 | 164K | $31.20 | 2025-09-22 | ||
| DeepSeek V3.1 DeepInfra, cheapest of 5 hosts | $0.25 | $0.95 | $0.13 | 164K | $29.25 | 2025-08-21 | |
| R1 0528 DeepInfra, cheapest of 4 hosts | $0.50 | $2.15 | $0.35 | 164K | $62.25 | 2025-05-28 | |
| DeepSeek V3 0324 SiliconFlow, cheapest of 2 hosts | $0.25 | $1.00 | 164K | $30.00 | 2025-03-24 | ||
| R1 served by Novita | $0.70 | $2.50 | 64K | $79.50 | 2025-01-20 | ||
| DeepSeek V3 StreamLake, cheapest of 2 hosts | $0.286 | $1.14 | 164K | $34.30 | 2024-12-26 |
12 of these 14 models are not served by DeepSeek itself on OpenRouter, so their price is the cheapest regular price among the hosts that run them, and the host is named under the model. Compare DeepSeek with every other provider
Common questions
How much does the DeepSeek API cost?
On 10 October 2026, DeepSeek models on this list cost from $0.0059 per million input tokens (DeepSeek V4 Flash 0731) to $0.70 (R1), and from $0.17 per million output tokens (DeepSeek V4 Flash 0423) to $3.48 (DeepSeek V4 Pro 0423).
What is the cheapest DeepSeek model?
For a typical mix of three input tokens to every output token, the cheapest is DeepSeek V4 Flash 0423 at $0.035 per million input tokens and $0.17 per million output tokens.
What is the newest DeepSeek model on the list?
DeepSeek V4.1 Flash, listed on 10 September 2026, at $0.15 per million input tokens and $0.60 per million output tokens.
Which DeepSeek model has the largest context window?
DeepSeek V4.1 Flash, with 1.05M tokens of context.
Does DeepSeek offer batch pricing?
Yes, 1 of its models on this list show a batch price, with batch input at a median of 75 percent of the standard price. Batch suits work that can wait for its answer.
Choosing a model for a production agent
Price is one input. Quality on your own cases, latency and the cost of retries decide the rest. Tell me what you are building.