>>
Run DeepSeek-V4.1-Flash-1017 on Lium
DeepSeek · Vision and text · 763.2B parameters · 510.3 GB 5-bit weights · released 2026-10-17
It needs about 613 GB of VRAM. The cheapest setup that runs it is 8x A100-SXM4-80GB (640 GB), from $13.44 per hour, billed per second.
Other GPUs that run it
How to run it
Rent the pod above, open a terminal over SSH or Jupyter and serve deepseek-ai/DeepSeek-V4.1-Flash-1017:
vllm serve deepseek-ai/DeepSeek-V4.1-Flash-1017 --tensor-parallel-size 8Claude Code and Codex can do all of it through the Lium CLI.
Frequently asked questions
How much VRAM does DeepSeek-V4.1-Flash-1017 need?
About 510.3 GB for the released 5-bit weights, so plan on 613 GB with room for activations and KV cache at a short context. Longer contexts and bigger batches need more.
What GPU should I rent to run DeepSeek-V4.1-Flash-1017?
8x A100-SXM4-80GB (640 GB). On Lium that setup costs from $13.44 per hour, billed per second with no minimum.
For AI agents (CLI install, agent skill, headless signup): llms-full.txt