NVIDIA Releases NemoClaw v0.0.122 for Autonomous Model Routing

NVIDIA Docs published release notes for NemoClaw v0.0.122, introducing an autonomous multi-model routing middleware layer for enterprise applications. The runtime dynamically routes prompts between Google Gemini, Anthropic Claude, OpenAI, and local DGX clusters based on real-time latency and token cost thresholds. This architecture insulates enterprise deployments against individual API outages, rate limit throttles, and sudden pricing shifts.