OrcaRouter Launches Orca Connect, Turning Any Local GPU into a Free, Globally Accessible AI API

New infrastructure bridges local and cloud AI inference, allowing developers to securely access, share, and scale their own GPU resources without port forwarding or complex configuration.

SAN FRANCISCO, Oct. 28, 2026 /PRNewswire/ -- Continuum AI today announced the launch of Orca Connect, a new feature on OrcaRouter that transforms locally hosted AI models into globally accessible APIs in seconds. Developers can now connect their personal GPUs, workstations, and on-premise servers to OrcaRouter's unified AI infrastructure, making local inference accessible from anywhere at zero platform cost.

As open-source AI models become increasingly powerful and consumer GPUs more capable, developers are running sophisticated models on their own hardware. However, accessing these models remotely, integrating them with AI agents, or sharing inference resources across teams has traditionally required complex networking, infrastructure management, and security configuration.

Orca Connect eliminates these barriers.

Supporting popular inference runtimes including Ollama, vLLM, LM Studio, llama.cpp, SGLang, MLX, and TensorRT-LLM, Orca Connect automatically discovers local models and connects them to OrcaRouter through a secure, outbound-only connection. No public IP addresses, port forwarding, or inbound firewall changes are required.

Key Capabilities

• Instant Global API Access: Turn locally running models into remotely accessible, OpenAI-compatible APIs using a single OrcaRouter API key.
• AI Agent Integration: Connect local GPUs directly to Claude Code, Codex, OpenCode, Gemini CLI, and other popular coding agents.
• Secure Model Sharing: Share selected models with teammates through revocable links with expiration dates and usage limits, without exposing API credentials.
• Hybrid Local-Cloud Inference: Combine local GPUs with OrcaRouter's 200+ hosted models, enabling optional cloud fallback when local resources are unavailable or overloaded.
• Zero Platform Fees: Inference served by users' own hardware is completely free through OrcaRouter. Only explicitly enabled hosted-model fallback incurs usage charges.

Orca Connect also introduces local-only API keys, configurable cloud fallback policies, and spending limits, giving developers precise control over where their models execute and when cloud resources may be used. Requests are relayed through OrcaRouter, while inference remains on the user's selected hardware unless cloud fallback is explicitly authorized.

Democratizing AI Infrastructure

Orca Connect represents another step toward OrcaRouter's vision of an open, decentralized AI infrastructure ecosystem, where developers retain ownership of their compute resources while benefiting from the accessibility and interoperability of cloud platforms.

Rather than forcing developers to choose between local inference and hosted AI services, OrcaRouter brings both into a single unified API.

Your GPU. Your cloud. Your rules.

Availability

Orca Connect is available now, free for all OrcaRouter users. Learn more: https://www.orcarouter.ai/solutions/orca-connect

About Continuum AI

Continuum AI is an AI research and infrastructure company developing OrcaRouter and OrcaCyber. Its mission is to make advanced artificial intelligence more accessible, secure, and open through unified inference infrastructure, intelligent routing, and specialized AI models.

Media Contact:

OrcaRouter Communications
+1 650-609-7501

[email protected] 
orcarouter.ai

SOURCE OrcaRouter