DeepSeek API pricing
Every DeepSeek model Inferray offers — 6 from DeepSeek — at its list price, with no gateway fee on top. One API key calls all of them, alongside every other model in the catalog.
DeepSeek models and prices
| Model | Input / 1M | Cached / 1M | Output / 1M | Context | Token Plans |
|---|---|---|---|---|---|
| DeepSeek V4.1 Flash CyberDeepSeek V4.1 Flash Cyber for OpenAI-style Chat Completions, Anthropic-style Messages, OpenAI-style Responses. | $0.3 | $0.007 | $1 | 1.0M | Pay-as-you-go |
| DeepSeek V4.1 Flash 1017DeepSeek V4.1 Flash 1017 for OpenAI-style Chat Completions, Anthropic-style Messages, OpenAI-style Responses. | $0.15 | $0.003 | $0.6 | 1M | Pay-as-you-go |
| DeepSeek V4.1 FlashReasoning and vision at a fraction of flagship prices. | $0.15 | $0.003 | $0.6 | 1M | Included |
| DeepSeek V4 FlashDeepSeek V4 Flash for OpenAI-style Chat Completions, Anthropic-style Messages, OpenAI-style Responses. | $0.14 | $0.0028 | $0.28 | 1.1M | Pay-as-you-go |
| DeepSeek V4 ProDeepSeek V4 Pro for OpenAI-style Chat Completions, Anthropic-style Messages, OpenAI-style Responses. | $0.66 | $0.022 | $1.98 | 1.1M | Pay-as-you-go |
| DeepSeek V3.2DeepSeek V3.2 for OpenAI-style Chat Completions, Anthropic-style Messages, OpenAI-style Responses. | $0.28 | $0.028 | $0.42 | 131K | Pay-as-you-go |
Which DeepSeek model should I use?
- DeepSeek V4.1 Flash Cyber is the newest, released October 29, 2026.
- DeepSeek V4 Flash costs the least per request: $0.14 in / $0.28 out per 1M tokens.
Each model page has worked cost examples, code, and the full spec.
Call the DeepSeek API
Point the OpenAI SDK (or the Anthropic SDK, or plain HTTP) at Inferray and name the model. This calls DeepSeek V4.1 Flash Cyber; swap the id for any model above.
from openai import OpenAI
client = OpenAI(base_url="https://api.inferray.com/v1", api_key="inferray_...")
reply = client.chat.completions.create(
model="deepseek-v4.1-flash-cyber",
messages=[{"role": "user", "content": "ship it"}],
)DeepSeek API questions
- How much does the DeepSeek API cost?
- On Inferray, DeepSeek models cost $0.14–$0.66 per 1M input tokens, the list price per model with no gateway fee on top. The table above has every model's input, cached input, and output price.
- Do I need a DeepSeek account to use DeepSeek?
- No. You call DeepSeek with your Inferray API key and pay from one Inferray balance or Token Plan, the same as for every other model in the catalog.
- Are DeepSeek models included in Token Plans?
- All of them except DeepSeek V4.1 Flash Cyber, DeepSeek V4.1 Flash 1017, DeepSeek V4 Flash, DeepSeek V4 Pro, and DeepSeek V3.2, which are pay-as-you-go only. Plan credit is spent at list price, so the 20 Solo plan gives you 60 of usage a month.
- Does my credit expire?
- No. Pay-as-you-go credit never expires, and a top-up credits exactly what you pay. Token Plan credit resets each month with the plan.
Other model families
Prices from the catalog as of November 12, 2026.