Model library / deepseek-ai
DeepSeek-V4-Pro-0813 is a 1650.5B mixture-of-experts open-weight model published by deepseek-ai on Hugging Face. It uses the DeepseekV4ForCausalLM architecture with 61 layers. The published configuration allows a context of 1,048,576 tokens. The weights are released under the mit licence.
6 of 384 experts active per token
Context length in tokens, as published
Weights at the released precision
Licence declared on the repository
Overview
DeepSeek-V4-Pro-0813 is a 1650.5B mixture-of-experts open-weight model published by deepseek-ai on Hugging Face. It uses the DeepseekV4ForCausalLM architecture with 61 layers. The published configuration allows a context of 1,048,576 tokens. The weights are released under the mit licence.
At the released precision the weights occupy about 1537 GiB, before any key-value cache. It therefore needs more than one accelerator, or a smaller precision. Only 6 of the 384 experts run per token, so compute per token is far below the total parameter count. The tables below work the memory out for common cards.
Specification
| Repository | deepseek-ai/DeepSeek-V4-Pro-0813 |
| Publisher | deepseek-ai |
| Published | 13 August 2026 |
| Architecture | DeepseekV4ForCausalLM |
| Model type | deepseek_v4 |
| Parameters | 1,650,497,936,906 |
| Experts | 384 total, 6 active per token |
| Layers | 61 |
| Hidden size | 7,168 |
| Attention heads | 128, 1 key/value heads |
| Context length | 1,048,576 tokens |
| Vocabulary | 129,280 tokens |
| Quantisation | fp8 |
| Licence | mit |
Memory
Parameter count multiplied by the bytes each format uses. The key-value cache comes on top.
| Precision | Weights | Notes |
|---|---|---|
| bf16 | 3074.3 GiB | Full precision as released |
| fp8 | 1537.1 GiB | 8-bit, near-lossless on most models |
| int4 | 768.6 GiB | 4-bit, smallest footprint |
The cache costs 122 KiB per token at 16-bit, so the full 1,048,576-token context of one request needs about 122.0 GiB. Concurrency multiplies that number, not the weights.
That figure is an upper bound: this model uses sliding-window attention with a 128-token window on some layers, so long requests cache less than this.
Hardware
At the released precision, with 90% of the card given to the server and 2.5 GiB kept for the runtime. The context column is what is left for the cache on that setup.
| GPU | Cards needed | Free for cache | Context it holds |
|---|---|---|---|
| RTX 4090 24GB | more than 8 | — | — |
| RTX 5090 32GB | more than 8 | — | — |
| L40S 48GB | more than 8 | — | — |
| A100 80GB | more than 8 | — | — |
| H100 80GB | more than 8 | — | — |
| RTX PRO 6000 96GB | more than 8 | — | — |
| H200 141GB | 8 | 163.8 GiB | the full 1,048,576 tokens |
Serving
A vLLM launch line for the shape above. Check the model card for a runtime the publisher recommends.
vllm serve deepseek-ai/DeepSeek-V4-Pro-0813 \ --tensor-parallel-size 8 \ --max-model-len 1048576 \ --gpu-memory-utilization 0.90
Questions
The weights take about 831 GiB at the released precision. No single card in the table above holds that, so it needs several GPUs or a lower precision.
The repository declares mit. Read the licence text before commercial use: the name alone does not tell you what is allowed.
The configuration allows 1,048,576 tokens. Whether the whole window is usable depends on the memory left for the cache, which the hardware table works out per card.
This page is a reference, and Aiqre does not serve this model today. We deploy open-weight models on dedicated EU hardware on request, so if you want an OpenAI-compatible endpoint for it, email [email protected] with the model name and roughly what volume you expect.
Also in the library
DeepSeek-V4-Flash-0731 is a 304.2 billion-parameter, mixture-of-experts language model with speculative decoding for agentic task…
Nex-N2.5-Max is a 1.6-trillion-parameter text-only MoE model designed for long-horizon agentic tasks.
Qwen3.8-2.4T-A95B is a 2.4 trillion-parameter causal language model with 95 billion active parameters, supporting up to 262 k tok…
Hosting
Tell us the model and roughly what volume you expect. We reply with a price and an OpenAI-compatible endpoint, with your prompts never stored.