⬡ Universal LLM Model Manager

One API for Every Model
Every Platform

Ollama · HuggingFace · GitHub — Unified

Reach is a Rust-native universal LLM model manager. It connects to three model platforms, handles acquiring and running models from each, and provides a single consistent API — regardless of where the model lives or how it runs. Self-hosted. No cloud dependency. No Python required.

Three Platforms. One Interface.

Reach treats each platform according to what it actually is — not a one-size-fits-all abstraction.

🦙

Ollama — Runtime Platform

Models already running locally. Reach calls Ollama's existing API to submit prompts and stream responses. No downloading, no loading — the model is already live.

🤗

HuggingFace — Weight Repository

Download GGUF and SafeTensors files from HuggingFace repos. Local storage with integrity verification via hash check after every download.

🐙

GitHub — Architecture Source

Clone repos, parse model configs, resolve weight locations. The most complex source — requires understanding repo structure before running anything.

Architecture

Caller → Unified API → Platform Router → Model Registry → Cleanup

[ Caller: OpenClaw / Any System ]
[ Reach Unified API ]
[ Platform Router ]
Ollama API
HuggingFace API
GitHub API
[ Model Registry ]
[ Cleanup Scheduler ]

Core Capabilities

What Reach actually does — no more, no less.

🔄

Platform Router

Inspects the incoming request, determines which platform the model lives on, and routes accordingly. The caller never needs to know where a model comes from.

📋

Model Registry

Tracks every model Reach has acquired — source, disk location, size, format, last access time, usage count. Persisted to ~/.reach/registry.json.

🧹

Cleanup Scheduler

Scans for stale models (default: 30 days). Reports file size reclaimed. Always dry-runs first — deletion requires explicit confirmation or --force.

Dash Integration

When Dash is present, model execution routes through Dash for CPU acceleration automatically. No configuration needed — Reach detects and uses it.

🔌

HTTP Server Mode

Optional REST API exposing Reach's full capability. OpenAI-compatible chat endpoint plus Reach-specific model management endpoints for non-Rust systems.

🦀

Rust-Native Library

Import directly as a Rust crate. Full async support. No Python, no wrappers, no cloud dependency. Self-contained binary.

What Reach Is Not

Clear boundaries. No feature creep.

Not an inference engine (that is Dash)
Not an accelerator (that is Dash)
Not tied to any single platform
Not a cloud service
Not a Python wrapper
Not an outreach platform

Get Reach

Rust-native. Self-hosted. One API for every model you'll ever need.

~ Source on GitHub