PawPaw the bear flexing, hyped about the community

Welcome to the Den

HiveBear is built by people who think AI should be for everyone - not just people with deep pockets and top-shelf hardware. Share your spare compute, contribute code, help someone get started, or just hang out. The pack looks after its own.

Top Benchmarked Models

Hardware Database

Real benchmarks from real machines, shared by the pack. Got a beat-up ThinkPad? Someone already figured out exactly what models run best on it.

No benchmarks found

Try adjusting the filters, or submit your own benchmark with the CLI.

Get involved

Discord

Hang out with the pack. Get help, share benchmarks, swap hardware war stories.

GitHub Discussions

Got an idea? A question? A wild feature request? The den is open.

Contributing Guide

First PR or fiftieth - there's room for every skill level. Come build with us.

Report an Issue

Found a bug? We want to hear about it. Seriously, you're helping.

Speak the language

A quick guide to HiveBear terminology.

🏠

The Hive

Where everyone chips in what they can. Your idle machine helps someone else run AI - and when you need a boost, the hive helps you right back.

🐻

Cubs

The little machines that need a hand - your old laptop, that Raspberry Pi, anything that can't quite run the big models alone. The pack raises the cubs.

🐾

Pack Members

That's you. Whether you share compute, write code, fix docs, or just help a newcomer in Discord - you're part of the pack.

✨

Beary Best Match

The AI model that fits your hardware like a glove. HiveBear sniffs it out so you don't have to guess.

Share your benchmark

Run hivebear benchmark and share your results. Someone with a machine just like yours will thank you for it.

Learn how
HiveBearHiveBear

Free, open-source, self-hosted AI that actually fits your machine. A P2P mesh of neighbors pooling everyday hardware to run big local AI models together. Written in Rust, powered by the hive.

Product

  • Download
  • Documentation
  • Playground
  • FAQ

Run a model

  • Run Llama 3 70B
  • Run DeepSeek R1
  • Run Qwen 2.5 72B
  • Run Mistral 7B
  • All models →

Compare

  • HiveBear vs Ollama
  • HiveBear vs LM Studio
  • HiveBear vs exo
  • HiveBear vs Jan.ai

Community

  • Discord
  • GitHub
  • Discussions
  • Community hub
PawPaw the bear, chilling

Built with Rust. MIT License. © 2026 BeckhamLabs.

Privacy Policy
HiveBearHiveBear
DownloadDocsModelsFAQCommunity
GitHubSign inInstall