D
DeepSeek V4 ProOPEN
DeepSeek V4 Pro (DeepSeek): live API pricing, context window and the best open-source alternatives, tracked daily by olud.ai.
Prices update automatically — checked hourly against provider list prices.
Benchmarks & performance
Independent benchmark scores for DeepSeek V4 Pro, measured by Artificial Analysis. Higher is better. Measurement mode: Reasoning, Max Effort.
For the same budget, DeepSeek V4 Pro delivers 13 times more capability. Open weights also mean you can self-host it and pay nothing per token.
About this model
DeepSeek V4 Pro is an open-weight AI model by DeepSeek. You can download and self-host it for free; the prices below are hosted-API list prices, tracked hourly, for when you prefer convenience over self-hosting.
Frequently asked questions
What is DeepSeek V4 Pro?
DeepSeek V4 Pro is an AI language model from DeepSeek. It is open-weight: you can download it and run it on your own hardware, for free. It scores 44.3 on the Artificial Analysis intelligence index.
Is DeepSeek V4 Pro free?
The weights are free and open — you can self-host DeepSeek V4 Pro and pay nothing per token. If you prefer a hosted API, list prices are $0.44 per million input tokens and $0.87 per million output tokens.
What is DeepSeek V4 Pro good at?
Independent benchmarks from Artificial Analysis give it GPQA 88.8%, Humanity's Last Exam 35.9%, Long Context Reasoning 66.3%, SciCode 50%, IFBench 76.5%, τ²-Bench 96.2%, τ-Bench Banking 25.8%, Terminal-Bench 64%, Terminal-Bench Hard 46.2%. It is particularly used for code generation.
How fast is DeepSeek V4 Pro?
It generates about 70.8 tokens per second, with a median 0.91s delay before the first token. Measured independently by Artificial Analysis.
Can I self-host DeepSeek V4 Pro?
Yes. DeepSeek V4 Pro has open weights, so you can download it and run it on your own GPU or server with tools like Ollama, vLLM or llama.cpp — with no per-token cost.