>>

Run ERNIE-4.5-300B-A47B-Paddle on Lium

Baidu · Text generation · 300.5B parameters · 601 GB 8-bit weights · released 2025-06-28
It needs about 722 GB of VRAM. The cheapest setup that runs it is 8x RTX PRO 6000 Server (768 GB), from $14.00 per hour, billed per second.

Other GPUs that run it

8x H2001128 GBfrom $34.80 / hourRent

How to run it

Rent the pod above, open a terminal over SSH or Jupyter and serve baidu/ERNIE-4.5-300B-A47B-Paddle:
vllm serve baidu/ERNIE-4.5-300B-A47B-Paddle --tensor-parallel-size 8
Claude Code and Codex can do all of it through the Lium CLI.

Frequently asked questions

How much VRAM does ERNIE-4.5-300B-A47B-Paddle need?

About 601 GB for the released 8-bit weights, so plan on 722 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 ERNIE-4.5-300B-A47B-Paddle?

8x RTX PRO 6000 Server (768 GB). On Lium that setup costs from $14.00 per hour, billed per second with no minimum.
All open models
For AI agents (CLI install, agent skill, headless signup): llms-full.txt