|
Quorom — Your private AI cluster
Quorom turns the machines you already own into one OpenAI-compatible inference cluster. Cross-platform desktop app, relay-routed networking, zero config.
|
85 % |
|
PicoCluster Claw — Desktop AI Datacenter
A two-node desktop AI datacenter: the OpenClaw automation runtime, dedicated CUDA inference, and the ThreadWeaver operator console — boxed, always-on, under...
|
84 % |
|
HostSSH — any hardware, fully portable, AI-ready
Onboard, backup, restore, and move any machine — cloud VPS, bare metal, office hardware, GPU rental. Deploy apps, run Files/DB/Workers/Storage, and expose an...
|
82 % |
|
dollama.net — Your computers, one API
Turn the computers you already own into a personal AI cluster. Use your own hardware from wherever you work. When your cluster needs more, draw on free capac...
|
82 % |
|
Locai — Orchestrate where your AI thinks
Locai is the intelligence, orchestration and routing layer for where your AI runs. One endpoint enforces your inference policy: routing every request to devi...
|
82 % |
|
Stedfast - Self-Hosted AI Interface
Self-hosted AI interface built for power users. Zero npm dependencies. One Docker command.
|
82 % |
|
Computed Layer — The compute layer for AI applications
Run inference across a global network of GPUs through a single OpenAI-compatible API. Dedicated GPUs from $0.29/hr. Deploy in under two minutes.
|
82 % |
|
Spider — One inference cloud, built from the world's idle GPUs
Spider aggregates idle GPU compute into a single managed inference cloud. The node executes. The cloud thinks.
|
82 % |
|
Gridline
Run GPU workloads on your own Vast.ai and RunPod accounts through one API, with beta host-replacement controls.
|
82 % |
|
Nethera - Deploy Docker Compose apps to any machine you own
Deploy Docker Compose apps to any machine you own and get a public HTTPS endpoint. No static IP, no port forwarding, no router config. One machine or a whole...
|
81 % |
|
Infervisor — one inference runtime for every modality
One inference runtime for every modality — LLMs, diffusion, and voice — rebuilt from first principles in Rust. Native on NVIDIA and AMD, on-prem or cloud.
|
81 % |
|
Blaze by E2E Networks
Blaze is a high-performance inference stack with intelligent routing — more intelligence per dollar, running on E2E Networks' own GPU infrastructure.
|
81 % |
|
Introduction | workflow-engine
workflow-engine is a TypeScript library that runs durable, multi-stage
|
81 % |
|
dinf — decentralized distributed inference
Run open models on hardware people already own. Share your Mac or GPU, earn credits, and use them for inference.
|
80 % |
|
spore.host — Ephemeral compute for researchers
Find the right EC2 instance, launch it in minutes, and let it manage itself. Auto-termination, idle detection, Slack control, and AI-assistant integration fo...
|
80 % |
|
osctrl — Open-source osquery fleet management
osctrl is an open-source osquery fleet manager for endpoint visibility, configuration, live queries, log collection, alerting, file carving and MCP-based AI...
|
80 % |
|
Nebari | Nebari
Enterprise-ready AI and Data Science Platform: Composable, Extensible, Open Source. Install notebooks, dashboards, and model serving on infrastructure you own.
|
80 % |
|
Valyent
A developers-first cloud platform.
|
80 % |
|
Decoupled AI — The Decentralized AI Network
Run a node. Share your compute. Access better AI models — privately, without centralized control.
|
80 % |
|
dxflow - Unifying Compute & Data Fleet Everywhere
Turn any machine into a node for running containerized workflows. dxflow gives you a unified CLI, REST API, and web console across Docker, Podman, Singularit...
|
80 % |
Detected social media networks