# mesh-llm

> Distributed AI/LLM for the people. Share compute privately or publicly to power your agents and chat.

- **URL**: https://www.freshcrate.ai/projects/mesh-llm
- **Author**: Mesh-LLM
- **Category**: Uncategorized
- **Latest version**: `v0.71.0` (2026-05-30)
- **License**: Apache-2.0
- **Source**: https://github.com/Mesh-LLM/mesh-llm
- **Homepage**: http://docs.anarchai.org/
- **Language**: Rust
- **GitHub**: 834 stars, 99 forks
- **Registry**: github (`Mesh-LLM/mesh-llm`)
- **Tags**: `agents`, `ai`, `decentralized`, `distributed`, `llm`, `rust`

## Description

Distributed AI/LLM for the people. Share compute privately or publicly to power your agents and chat.

## Recent releases

| Version | Date | Urgency | Changes |
| --- | --- | --- | --- |
| `v0.71.0` | 2026-05-30 | High | Latest release: v0.71.0 |
| `v0.66.0` | 2026-05-22 | High | Latest release: v0.66.0 |
| `v0.65.1` | 2026-05-05 | High | ## What's Changed * Preflight Vulkan launches against primary GPU VRAM by @i386 in https://github.com/Mesh-LLM/mesh-llm/pull/385 * Revalidate sticky auto routing for media requests by @IvGolovach in https://github.com/Mesh-LLM/mesh-llm/pull/404 * Expose ranked model targets in the API by @IvGolovach in https://github.com/Mesh-LLM/mesh-llm/pull/403 * Prepare repo layout for skippy merges by @i386 in https://github.com/Mesh-LLM/mesh-llm/pull/420 * feat: add built-in pi launcher (mesh-llm pi) by @m |
| `v0.65.0` | 2026-05-01 | High | ## What's Changed * chore(ci): build SDK stuff when Rust / FFI changes by @ndizazzo in https://github.com/Mesh-LLM/mesh-llm/pull/353 * Add per-model `parallel` config to control the number of parallel request slots per model by @ndizazzo in https://github.com/Mesh-LLM/mesh-llm/pull/357 * Fix startup on nodes with pinned Tegra-based GPU SoCs by @ndizazzo in https://github.com/Mesh-LLM/mesh-llm/pull/360 * Bump iroh to 0.98, move legacy Fly relay to tools/ by @michaelneale in https://github.com/Mes |
| `v0.64.0` | 2026-04-20 | High | ## What's Changed * fix: remove unintended PR job from 307 by @ndizazzo in https://github.com/Mesh-LLM/mesh-llm/pull/310 * Share split GGUF MoE rankings under stable distribution refs by @i386 in https://github.com/Mesh-LLM/mesh-llm/pull/314 * feat: Implement light mode for topology diagram by @ndizazzo in https://github.com/Mesh-LLM/mesh-llm/pull/302 * [codex] Bump llama.cpp fork pin to e88186e78777 by @i386 in https://github.com/Mesh-LLM/mesh-llm/pull/315 * Use published MoE analysis for fit p |
| `v0.63.0-rc5` | 2026-04-18 | High | **Full Changelog**: https://github.com/Mesh-LLM/mesh-llm/compare/v0.63.0-rc4...v0.63.0-rc5 |
| `v0.63.0-rc4` | 2026-04-18 | High | ## What's Changed * Run releases from GitHub Actions instead of tag pushes by @i386 in https://github.com/Mesh-LLM/mesh-llm/pull/323 * feature: add embedded SDK support by @ndizazzo in https://github.com/Mesh-LLM/mesh-llm/pull/233 * Let moe analyze share rankings automatically by @i386 in https://github.com/Mesh-LLM/mesh-llm/pull/325 * Add optional CUDA and ROCm skip for prerelease releases by @i386 in https://github.com/Mesh-LLM/mesh-llm/pull/326 * Keep auto-routed chats on one model and one pe |
| `v0.63.0-rc2` | 2026-04-18 | High | ## What's Changed * Update LLAMA_CPP_SHA hash value by @i386 in https://github.com/Mesh-LLM/mesh-llm/pull/320 * models: inspect and clean up mesh-managed cache entries safely by @IvGolovach in https://github.com/Mesh-LLM/mesh-llm/pull/300 * Align MoE split cache naming with llama.cpp by @i386 in https://github.com/Mesh-LLM/mesh-llm/pull/321   **Full Changelog**: https://github.com/Mesh-LLM/mesh-llm/compare/v0.63.0-rc.1...v0.63.0-rc2 |
| `v0.63.0-rc.1` | 2026-04-17 | High | ## What's Changed * fix: remove unintended PR job from 307 by @ndizazzo in https://github.com/Mesh-LLM/mesh-llm/pull/310 * Share split GGUF MoE rankings under stable distribution refs by @i386 in https://github.com/Mesh-LLM/mesh-llm/pull/314 * feat: Implement light mode for topology diagram by @ndizazzo in https://github.com/Mesh-LLM/mesh-llm/pull/302 * [codex] Bump llama.cpp fork pin to e88186e78777 by @i386 in https://github.com/Mesh-LLM/mesh-llm/pull/315 * Use published MoE analysis for fit p |
| `v0.62.1` | 2026-04-17 | High | ## What's Changed * Require mesh-llm/1 for mesh joins by @i386 in https://github.com/Mesh-LLM/mesh-llm/pull/299 * scripts: fix silent build exit when neither ccache nor sccache is installed by @ventz in https://github.com/Mesh-LLM/mesh-llm/pull/305 * remove cuda 103 pr259 by @ndizazzo in https://github.com/Mesh-LLM/mesh-llm/pull/307  ## New Contributors * @ventz made their first contribution in https://github.com/Mesh-LLM/mesh-llm/pull/305  **Full Changelog**: https://github.com/Mesh-LLM/mesh-ll |

## Citation

- HTML: https://www.freshcrate.ai/projects/mesh-llm
- Markdown: https://www.freshcrate.ai/projects/mesh-llm.md
- Dependencies JSON: https://www.freshcrate.ai/api/projects/mesh-llm/deps

_Generated by freshcrate.ai. Indexes github releases for AI-agent ecosystem packages._