deepseek
DeepSeek: DeepSeek V4 Pro 0423
deepseek/deepseek-v4-proInput: $0.951432 /M tokens
Model description
An overview of the model’s capabilities and characteristics.
DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window. It is designed for advanced reasoning, coding, and long-horizon agent workflows, with strong performance across knowledge, math, and software engineering benchmarks.
Built on the same architecture as DeepSeek V4 Flash, it introduces a hybrid attention system for efficient long-context processing. Reasoning efforts high and xhigh are supported; xhigh maps to max reasoning. It is well suited for complex workloads such as full-codebase analysis, multi-step automation, and large-scale information synthesis, where both capability and efficiency are critical
- Developer
- deepseek
- Model ID
- deepseek/deepseek-v4-pro
- Modality
- text → text
- Context
- 1,048,576
- Maximum response
- 393,216
Prices
Tariff confirmed: 2026-09-16 14:43 UTC
Verified official rate
| Rate | USD | Unit |
|---|---|---|
| Input | $0.951432≈ 11,203 UZS | /M tokens |
| Output | $1.902864≈ 22,406 UZS | /M tokens |
| Cache read | $0.079286≈ 934 UZS | /M tokens |
CBU rate, Sep 16, 2026. Prices are shown in USD with a reference equivalent in UZS. The bank rate used for payment may differ.
API
Check the available parameters and actual API contract in the documentation before making a request.
API documentationSupported parameters
frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_completion_tokensmax_tokensmin_ppresence_penaltyreasoningreasoning_effortrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_pFAQ
Where can I check API integration?
Check the available parameters and request contract in the API documentation. API documentation
Where are model and data terms described?
Execution and data-processing terms are described in cicora.ai materials. Execution and model terms · Privacy
What can this model accept and generate?
Input: text. Output: text.
Which limits are listed?
Context: 1,048,576. Maximum response: 393,216.
Which request parameters are listed?
The saved record lists: frequency_penalty, include_reasoning, logit_bias, logprobs, max_completion_tokens, max_tokens, min_p, presence_penalty, reasoning, reasoning_effort, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs, top_p.
How is the rate shown?
The verified official rate appears above. Units: /M tokens.
Technical details
- Model ID
- deepseek/deepseek-v4-pro
- Modality
- text->text
- Context
- 1,048,576
- Supported parameters
frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_completion_tokensmax_tokensmin_ppresence_penaltyreasoningreasoning_effortrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p- Tariff confirmed
- 2026-09-16 14:43 UTC
- Tariff conditions
- Input · /M tokensOutput · /M tokensCache read · /M tokens