DeepSeek API pricing
Every DeepSeek model we track, with the cheapest price per 1M tokens across 24 providers next to DeepSeek's own API price. Collected daily.
DeepSeek API prices by model
Standard (real-time) rates per 1M tokens, newest first. “Cheapest” is the lowest input price across every provider we track; batch and cached-input rates are on each model page.
| Model | Cheapest input | Output | Cheapest on |
|---|---|---|---|
| DeepSeek V4.1 Flash | $0.150 | $0.600 | Heabsy |
| DeepSeek V4 Flash1.0M context | $0.0099 | $1.28 | OpenRouter |
| DeepSeek V4 Pro1.0M context | $0.435 | $0.870 | Velokey |
| DeepSeek V4 Flash Vision | $0.216 | $0.647 | OpenRouter |
| DeepSeek V3.2 | $0.209 | $0.310 | GMI Cloud |
| DeepSeek V3.133K context | $0.250 | $0.950 | Deep Infra |
| DeepSeek V3.1 Terminus | $0.270 | $1.00 | Prime Intellect |
| DeepSeek V364K context | $0.290 | $1.14 | GMI Cloud |
| DeepSeek R1 0528 Qwen3 8B | $0.060 | $0.090 | Novita AI |
| DeepSeek R1 Distill Qwen 1.5B | $0.180 | $0.180 | Together AI |
| DeepSeek R164K context | $0.500 | $2.15 | Deep Infra |
| DeepSeek R1 Distill Llama 70B | $2.00 | $2.00 | Together AI |
| DeepSeek R1 Distill Qwen 14B | $1.60 | $1.60 | Together AI |
| DeepSeek Coder 33B16K context | $0.800 | $0.800 | Together AI |
Official column shows input / output per 1M tokens on DeepSeek's own API.
DeepSeek price history
Average input price per 1M tokens across the providers serving each model, last 90 days.
Frequently Asked Questions
What is the cheapest DeepSeek API?
Of the 14 DeepSeek models we track, DeepSeek V4 Flash has the lowest input price right now: $0.0099 per 1M input tokens on OpenRouter. Prices are collected daily from 24 providers — see the table above for every model.
Where is DeepSeek V4.1 Flash cheapest?
DeepSeek V4.1 Flash is served by 16 providers we track. The lowest standard input price is $0.150 per 1M tokens on Heabsy. Throughput, latency and context limits differ between hosts, so check the model page before committing.
Is a third-party host cheaper than DeepSeek's own API?
For 2 of the 14 DeepSeek models we track, another provider currently lists a lower input price than DeepSeek's own API. Compare the "Cheapest" and "Official" columns above; batch and cached-input rates are on each model page.