AI chat and search
Ask follow-up questions and use intelligent search in the DeepSeek assistant.
DeepSeek
Freemium
Chat, search for information, and build AI applications with reasoning and image-capable models.
DeepSeekとは？
DeepSeek combines an AI assistant with a developer model API. Use its browser chat or mobile app to ask questions, work through problems, and search for information. Developers can integrate DeepSeek models into applications through OpenAI-compatible or Anthropic-compatible interfaces. DeepSeek-V4.1-Flash also accepts images for tasks such as describing pictures, reading screenshot text, and interpreting charts.
DeepSeekでできること
Optional reasoning
Choose thinking or non-thinking modes when calling models through the API.
Image understanding
Send images to the Flash API to describe their contents, read text, or analyze charts.
Application integration
Connect applications using OpenAI-compatible or Anthropic-compatible API formats and an API key.
Structured and tool-based workflows
Use JSON output and tool calls to connect model responses with application logic.
利用シーン
“Explore a question”
Discuss a topic with the assistant, ask follow-up questions, and search for relevant information.
“Interpret visual material”
Build a workflow that asks the Flash model about screenshots, diagrams, or charts supplied as images.
“Add AI to an application”
Combine reasoning, structured responses, and tool calls with your application’s own data and functions.
よくある質問
How can I use DeepSeek?
Use the browser chat or the official iOS and Android assistant. Developers can create an API key to integrate model calls into their own applications.
Is the mobile assistant free?
The official DeepSeek mobile assistant offers free interaction with its AI models. Developer API usage is billed separately by token consumption.
What does the Flash API cost?
For deepseek-flash, off-peak rates per one million tokens are $0.003 for cached input, $0.15 for uncached input, and $0.60 for output. Peak rates are $0.006, $0.30, and $1.20 respectively. Peak hours are 01:00–04:00 and 06:00–10:00 UTC, Monday through Friday, excluding Chinese public holidays; all other hours are off-peak. Rates apply to API usage and depend on the selected model and cache status.
Which model supports image input?
DeepSeek-V4.1-Flash, called through the deepseek-flash API model name, accepts images alongside text. It can describe pictures, read screenshot text, and analyze charts.
DeepSeek のトラフィック分析
トラフィックデータは参考情報としてご利用ください。
- ドメイン評価
- 90
DeepSeekの代替品
ChatGPT
chatgpt.com
A conversational AI assistant for writing, research, learning, planning, coding, and file-based tasks.
Claude
claude.ai
Anthropic’s AI assistant for writing, coding, document analysis and cited research.
Google Gemini
gemini.google.com
Write, research, analyze files and talk through ideas with Google’s AI assistant
Doubao
doubao.com
An AI assistant for questions, writing, voice conversations and visual creation, with a desktop work mode.
Qwen Studio
chat.qwen.ai
Use Qwen’s AI assistant for conversations, research, visual creation, and webpage generation.