>>

Run Kimi-Dev-72B on Lium

Moonshot AI · Text generation · 72.7B parameters · 145.4 GB BF16 weights · released 2025-06-16
It needs about 175 GB of VRAM. The cheapest setup that runs it is 4x L40S (192 GB), from $1.52 per hour, billed per second.

Other GPUs that run it

4x RTX A6000192 GBfrom $1.68 / hourRent
8x RTX 5090256 GBfrom $4.80 / hourRent

How to run it

Rent the pod above, open a terminal over SSH or Jupyter and serve moonshotai/Kimi-Dev-72B:
vllm serve moonshotai/Kimi-Dev-72B --tensor-parallel-size 4
Claude Code and Codex can do all of it through the Lium CLI.

Frequently asked questions

How much VRAM does Kimi-Dev-72B need?

About 145.4 GB for the released BF16 weights, so plan on 175 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 Kimi-Dev-72B?

4x L40S (192 GB). On Lium that setup costs from $1.52 per hour, billed per second with no minimum.
All open models
For AI agents (CLI install, agent skill, headless signup): llms-full.txt