Starstruk Logo MarkStarstruk

v0.56.0

Mesh-LLM/mesh-llmv0.56.0Apr 7, 2026by github-actions[bot]

AI Summary

A major release featuring comprehensive plugin system support, enhanced auto-update flows, and significant cryptographic improvements.

Key Highlights

  • New plugin system for MCP, HTTP, and inference endpoints
  • Explicit update command and opt-in auto-update flow
  • Crypto primitives for owner keystore and signed/encrypted messages
  • Added multimodal blob uploads and console support
  • Auto-enable row split for local multi-GPU (tensor parallelism)

New Features

  • Plugin system (MCP, HTTP, inference)
  • Explicit update command
  • Multimodal blob uploads
  • Auto-enable row split for multi-GPU
  • Crypto primitives (keystore, signed messages)

Full Release Notes

## What's Changed
* Disable windows by @i386 in https://github.com/michaelneale/mesh-llm/pull/143
* Remove client nodes from topology visualization by @i386 in https://github.com/michaelneale/mesh-llm/pull/145
* feature: mesh topology selector by @ndizazzo in https://github.com/michaelneale/mesh-llm/pull/152
* add navigable node detail sidebar by @i386 in https://github.com/michaelneale/mesh-llm/pull/148
* [codex] make README friendlier and split long reference docs by @i386 in https://github.com/michaelneale/mesh-llm/pull/155
* Add workflow to clean up PR GitHub Actions caches on close by @i386 in https://github.com/michaelneale/mesh-llm/pull/158
* Add GitHub cache versioning and a full cache reset workflow by @i386 in https://github.com/michaelneale/mesh-llm/pull/159
* update hf-hub to checksum verification rev by @i386 in https://github.com/michaelneale/mesh-llm/pull/154
* Update mesh-llm crate README to match current architecture by @i386 in https://github.com/michaelneale/mesh-llm/pull/157
* Add explicit update command and opt-in auto-update flow by @i386 in https://github.com/michaelneale/mesh-llm/pull/138
* fix: --no-draft flag ignored + case-sensitive cache path mismatch by @vveerrgg in https://github.com/michaelneale/mesh-llm/pull/160
* Add multimodal blob uploads and console support by @i386 in https://github.com/michaelneale/mesh-llm/pull/153
* Deflake CI smoke tests by @michaelneale in https://github.com/michaelneale/mesh-llm/pull/167
* Fix llama-server failing to start on single-node, update llama.cpp fork to upstream by @michaelneale in https://github.com/michaelneale/mesh-llm/pull/172
* [codex] Namespace server logs by mesh-llm PID by @i386 in https://github.com/michaelneale/mesh-llm/pull/173
* test(ui): add selective PR-130 regressions and HF cache matching by @ndizazzo in https://github.com/michaelneale/mesh-llm/pull/162
* [codex] Sync owner keystore crypto with main by @i386 in https://github.com/michaelneale/mesh-llm/pull/178
* crypto: owner keystore and signed+encrypted message primitives by @Bortlesboat in https://github.com/michaelneale/mesh-llm/pull/156
* feat: auto-enable row split for local multi-GPU (tensor parallelism) by @michaelneale in https://github.com/michaelneale/mesh-llm/pull/166
* Build the new plugin system for MCP, HTTP, and inference endpoints by @i386 in https://github.com/michaelneale/mesh-llm/pull/169
* Resolve managed models from the Hugging Face cache only by @i386 in https://github.com/michaelneale/mesh-llm/pull/175
* Add serve/client commands and local GPU inspection for #184 by @i386 in https://github.com/michaelneale/mesh-llm/pull/185
* local config support on top of #185 by @i386 in https://github.com/michaelneale/mesh-llm/pull/186
* Automatic MoE expert ranking improvements by @i386 in https://github.com/michaelneale/mesh-llm/pull/137
* [codex] Document Lemonade integration by @i386 in https://github.com/michaelneale/mesh-llm/pull/190
* Use OS keychain for storing identity keystore by @i386 in https://github.com/michaelneale/mesh-llm/pull/187
* [codex] Keep /api/status off model inventory by @i386 in https://github.com/michaelneale/mesh-llm/pull/192
* feature + fix: fixed layout bug, node selection tooltip bug, added table selection by @ndizazzo in https://github.com/michaelneale/mesh-llm/pull/198
* Remove deprecated models migrate command by @i386 in https://github.com/michaelneale/mesh-llm/pull/200
* fix: hard crash when splitting KV quantizations with flash attention by @ndizazzo in https://github.com/michaelneale/mesh-llm/pull/199
* feat(protocol): add distributed config protocol with ownership by @ndizazzo in https://github.com/michaelneale/mesh-llm/pull/193
* Make release version bump fail fast and update all tracked manifests by @i386 in https://github.com/michaelneale/mesh-llm/pull/201

## New Contributors
* @vveerrgg made their first contribution in https://github.com/michaelneale/mesh-llm/pull/160
* @Bortlesboat made their first contribution in https://github.com/michaelneale/mesh-llm/pull/156

**Full Changelog**: https://github.com/michaelneale/mesh-llm/compare/v0.55.1...v0.56.0