
DeepSeek API
DeepSeek V and R models from DeepSeek, through one OpenAI-compatible API.
DeepSeek builds open-weight chat and reasoning models with some of the lowest prices per token for their quality. Use them through UltraGPT with the OpenAI SDK, paying per token from one balance.
Models
6
From / 1M input
$0.019
Deepseek V4 Flash
Largest context
1.0M
DeepSeek V4.1 Flash
DeepSeek models and prices
Chat and embedding prices are USD per 1M tokens: input / output.
| Model | Type | Context | Price |
|---|---|---|---|
DeepSeek V4.1 Flashdeepseek-v4.1-flash | Chat | 1.0M | $0.3 / $1.2 |
DeepSeek V4.1 Flash 1017deepseek-v4.1-flash-1017 | Chat | 1.0M | $0.036 / $0.5 |
DeepSeek V4 Prodeepseek-v4-pro | Chat | 1.0M | $0.9483 / $1.897 |
Deepseek V4 Flashdeepseek-v4-flash | Chat | 1.0M | $0.019 / $1.28 |
DeepSeek V4 Flash Vision Expdeepseek-v4-flash-vision-exp | Chat | 1.0M | $0.44 / $1.32 |
DeepSeek V3.2deepseek-v3.2 | Chat | 164K | $0.26 / $0.38 |
Call DeepSeek V4.1 Flash in a few lines
Swap the model id for any model in the table above; nothing else changes.
curl https://api.ultragpt.pro/v1/chat/completions \
-H "Authorization: Bearer $ULTRAGPT_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek-v4.1-flash",
"messages": [{ "role": "user", "content": "Hello!" }]
}'Comparisons
FAQ
How do I get a DeepSeek API key?
Sign in to UltraGPT Developers, top up your balance and create an API key. The same key works for every DeepSeek model and every other model on UltraGPT - no separate DeepSeek account needed.
How much does the DeepSeek API cost on UltraGPT?
Chat models of DeepSeek start at $0.019 per 1M input tokens (Deepseek V4 Flash). You pay per token from a prepaid balance, with no subscription; each model's page lists its exact input, cached-input and output prices.
Does it work with the official SDK?
UltraGPT is OpenAI-compatible: use the official OpenAI SDK (Python, Node.js) or any OpenAI-compatible tool with the base URL https://api.ultragpt.pro/v1 and the model id from the table.
Which DeepSeek model has the largest context window?
DeepSeek V4.1 Flash (deepseek-v4.1-flash), with 1,048,576 tokens.