D
DeepSeek V3 0324OPEN
DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iteration of the flagship chat model family from the DeepSeek team.
Prices update automatically — checked daily against provider list prices.
- $0.44 per 1M tokens, assuming 3 input for 1 output the convention used by Artificial Analysis
- cheaper than 43% of the 191 open-weight models tracked here
- 6 of 14 DeepSeek models tracked here, from cheapest
Benchmarks & performance
Independent benchmark scores for DeepSeek V3 0324, measured by Artificial Analysis. Higher is better.
For the same budget, DeepSeek V3 0324 delivers 5 times more capability. Open weights also mean you can self-host it and pay nothing per token.
Models at this level cost $0.45 per 1M tokens (median of 20) — this one costs $0.44.
Cheaper and better on this index: GLM 5.3 Flash · Gemini 3.7 Flash (batch) · GPT-5.6 Luna (batch) and 32 more
About this model
DeepSeek V3 0324 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 daily, for when you prefer convenience over self-hosting.
Frequently asked questions
What is DeepSeek V3 0324?
DeepSeek V3 0324 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 15.2 on the Artificial Analysis intelligence index.
Is DeepSeek V3 0324 free?
The weights are free and open — you can self-host DeepSeek V3 0324 and pay nothing per token. If you prefer a hosted API, list prices are $0.25 per million input tokens and $1 per million output tokens.
What is DeepSeek V3 0324 good at?
Independent benchmarks from Artificial Analysis give it GPQA 65.5%, MMLU-Pro 81.9%, Humanity's Last Exam 4.7%, Long Context Reasoning 41.3%, LiveCodeBench 40.5%, SciCode 35.8%, MATH-500 94.2%, AIME 52%, AIME 2025 41%, IFBench 41%, τ²-Bench 47.1%, τ-Bench Banking 4.7%, Terminal-Bench 13.9%, Terminal-Bench Hard 15.2%. It is particularly used for code generation.
Can I self-host DeepSeek V3 0324?
Yes. DeepSeek V3 0324 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.