>>
Run Qwen3.8-2.4T-A95B on Lium
Alibaba Qwen · Text generation · 2.4T parameters · 4892.4 GB BF16 weights · released 2026-08-08
It needs about 5871 GB of VRAM, more than one 8-GPU node.
How to run it
Rent the pod above, open a terminal over SSH or Jupyter and serve Qwen/Qwen3.8-2.4T-A95B:
vllm serve Qwen/Qwen3.8-2.4T-A95BClaude Code and Codex can do all of it through the Lium CLI.
Frequently asked questions
How much VRAM does Qwen3.8-2.4T-A95B need?
About 4892.4 GB for the released BF16 weights, so plan on 5871 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 Qwen3.8-2.4T-A95B?
It needs more than one 8-GPU node at its released precision. Request a multi-node cluster on Lium, or run a quantized version.
For AI agents (CLI install, agent skill, headless signup): llms-full.txt