DeepSeek V4 Pro 0813
Neural Network
DeepSeek V4 Pro 0813 is a large mixture-of-experts model from DeepSeek, available in BotHub via a unified DeepSeek API.
Specifications
- Input
- text, images, documents
- Output
- text
- Context size
- 1 048 576 tokens
- Max answer length
- 943 718 tokens
- Reasoning effort
- none, low, high, max
- Capabilities
- reasoning, function calling
Prompt cost
(per 1M tokens)
Answer cost
(per 1M tokens)
How it works DeepSeek V4 Pro 0813?
DeepSeek V4 Pro 0813 is a large language model from DeepSeek based on the mixture-of-experts architecture: not all parameters process the request at once, but selected experts do. The 0813 index denotes a specific build, and the release itself is a public, stable version of the V4 Pro line, not a preview. In BotHub, you can use it without a VPN or foreign card: pay with Russian cards in rubles based on usage; tokens do not expire, and no subscription is needed. Over 250 models are available in one window, making it easy to compare answers and switch models without rewriting integrations. Product teams will appreciate the unified OpenAI-compatible API, AES-GCM encrypted chats, and corporate options like contracts, invoices, electronic document management, and an admin panel with limits. Typical scenarios: drafting and editing text, analyzing long documents and summarizing, coding and debugging assistance, answering questions within corporate services, and API-based assistant prototypes.Frequently asked questions about DeepSeek V4 Pro 0813
You can use generated results for commercial purposes. You own all rights to the content you create. The only restriction: make sure your prompt does not include copyrighted third-party material. You are responsible for respecting the rights to any input data.
deepseek-v4-pro-0813 is a text model with reasoning capabilities: it analyzes long documents, writes and refactors code, finds bugs, and solves logic and STEM problems. It accepts both text and images/documents as input and can call external functions. Suitable for analytics, coding, and agentic scenarios.
The single response limit is 393,216 tokens, meaning the model can output a very large amount of text at once: a long document, a detailed analysis, or a large block of code. The context window is 1,048,576 tokens, so it can accommodate extensive source material.
Yes, reasoning before answering is confirmed for deepseek-v4-pro-0813: the model breaks down the task step-by-step before formulating the result. This helps with math, logic, code debugging, and situations where it is important to arrive at the answer sequentially rather than guessing.
Yes, both. The model supports function calling and structured JSON output, making it easy to integrate into agents and workflows. In BotHub, it is available via a unified OpenAI-compatible API: you can switch models without rewriting your integration.
Images and documents — yes, you can send them along with text: the model reads the content and answers based on it. Audio and video are not supported in our data, so do not count on them. Generation of anything other than text is also not claimed for this model.
The quality of Russian in this model has not been verified in our data, so we cannot promise precise phrasing — it is easier to test it with your own task. The BotHub interface is in Russian, payment is in rubles, access is without VPN or foreign cards, and you can always compare the answer with another model.