Kimi K3 on a 16x GB10 cluster
The full Kimi K3 model across 16 NVIDIA GB10 nodes at 20+ tok/s.
Creator credited on the original post (see below).
Build sheet
- Parts
- 16x NVIDIA GB10 cluster
- AI model or service
- Kimi K3 local LLM inference
- Difficulty
- Advanced
Overview
Runs the full Kimi K3 model on a 16-node NVIDIA GB10 cluster at 20+ tok/s.
How it works
The full Kimi K3 model runs locally on a cluster of 16 NVIDIA GB10 nodes.
Hardware
- 16x NVIDIA GB10 cluster
Like this build? Get the weekly builds
Get the best new AI hardware builds every Thursday. One email a week, free.
Built something? Submit a build
This page is a short summary written by HardwareLab. Build steps, code and photos belong to the creator; follow “View original” for the full project.
Featured? Add this badge
Built this? Paste the badge into your README. It links back to this page.
Markdown
[](https://hardwarelab.dev/projects/kimi-k3-gb10-cluster/?utm_source=badge&utm_medium=referral&utm_campaign=launch_2w)HTML
<a href="https://hardwarelab.dev/projects/kimi-k3-gb10-cluster/?utm_source=badge&utm_medium=referral&utm_campaign=launch_2w"><img src="https://hardwarelab.dev/badge.svg" alt="Featured on HardwareLab" width="188" height="20"></a>Related builds
All 22 in Local LLM Rigs & Edge Agents
Muse Gadgets SDK
Meta’s open SDKs for connecting Muse to ESP32 and Raspberry Pi gadgets.
- ESP32 boards
- Raspberry Pi (3B+, 4, 5 or Zero 2 W) or another Linux computer with Bluetooth LE
Intermediate2.2K
ESP32 iPhone Agent
An ESP32-S3 board lets an AI model use a real iPhone: no computer at run time, no jailbreak.
- Seeed XIAO ESP32S3 (plain, 8MB PSRAM; any ESP32-S3 with native USB + PSRAM)
- iPhone on iOS 18+ with cellular data
- Lightning to USB Camera Adapter (Lightning iPhones only)
- +2 more
Intermediate2.4K