While AI Agent services are useful, most require users to prepare their own host running 24/7 and have a certain technical barrier to entry. The development team behind the open-source AI agent framework Hermes Agent, Nous Research, officially announced today that Hermes Agent can now be deployed with one click through the Nous Portal cloud platform. Users only need to select the model and server specifications, and in about 60 seconds, they can launch an AI agent running 24/7 without managing any infrastructure themselves.
Previously, users who wanted to run Hermes Agent had to install the CLI themselves, configure model providers, set up terminal access, and manage their own infrastructure. Many beginners in the community reported that Docker setup and API key management alone took days or even weeks. Now, all these complex setup steps are handled by the cloud platform, significantly lowering the barrier to entry.
Two Steps, 60 Seconds: Hermes Agent Goes Cloud
According to Nous Research’s announcement on X, the setup process for Hermes Agent Cloud is extremely simplified: choose a model, choose a server size, and after just two steps, you get a working AI agent in approximately 60 seconds. This agent features persistent memory capabilities, supports multiple communication platforms including Telegram, Discord, Slack, WhatsApp, and web chat, and runs in an isolated sandbox environment for security.
Hermes Agent is now in the Cloud!
Setup couldn’t be simpler: pick a model and a server size. Two clicks and 60 seconds later, your agent is live.
Running a team? Spin up agents for everyone at your org with granular access controls and unified billing, all from Nous Portal. pic.twitter.com/WrMRnODliV
— Nous Research (@NousResearch) July 8, 2026
For team users, Hermes Agent Cloud also provides fine-grained access control and unified billing. Administrators can create separate agents for different members within the organization, each with its own settings and memory space, with all charges consolidated into a single invoice—no need to apply for separate API keys or manage individual cloud hosts for each member.
The cloud agent and self-hosted versions offer nearly identical functionality, including full toolchain support: web search and crawling, image generation, voice synthesis, code execution, scheduled tasks, and more. The only difference is that the hardware is managed by Nous Research, so users don’t need to worry about server maintenance, Docker updates, or security patches (essentially, it’s like renting a hosted server).
Nous Portal: Liberation from API Key Hell
The core foundation of Hermes Agent Cloud is Nous Portal, which was launched by Nous Research in April 2026.Unified Subscription PlatformPortal integrates over 200 AI models, covering Anthropic’s Claude series (Fable 5, Opus 4.7, Sonnet 4.6, Haiku 4.5), OpenAI’s GPT series (GPT-5.5, GPT-5.4 Mini), Google’s Gemini series (Gemini 3 Pro, Gemini 3 Flash), as well as open-source models like DeepSeek, Llama, Mistral, and Qwen.
Beyond model access, Portal also comes equipped with a complete Tool Gateway, integrating five commonly used agent tools under a single subscription:
- Web Search and CrawlingPowered by Firecrawl, no API key or rate limit management needed.
- Image generationSupports nine models including FLUX 2 Klein 9B, FLUX 2 Pro, GPT Image, Ideogram V3, and Recraft V4 Pro.
- Speech synthesisHigh-quality TTS without needing a separate OpenAI Key.
- Browser AutomationPowered by Browser Use, enabling agents to operate real browsers.
- Code Execution SandboxOptional serverless terminal sandbox for secure code execution.
Before Portal was available, users needed to separately register for multiple services including OpenRouter, Anthropic, OpenAI, Firecrawl, FAL, and Browser Use to get Hermes Agent fully operational, managing their own API keys and billing for each. Portal consolidates all of this into a single OAuth login and one bill, significantly simplifying the management overhead. This improvement is particularly important for Windows users, as configuring API keys for each tool individually in the Windows environment has been one of the most challenging aspects of the entire Hermes installation process.
Cloud vs. Self-Hosting: Which One Is Right for You?
The Hermes Agent open-source project itself remains completely free, allowing users to continue choosing self-hosting and using any model provider they prefer, including OpenRouter, Anthropic, direct OpenAI API, local Ollama, and more. The cloud version is a paid service designed for users who don’t want to manage infrastructure, offering a “plug-and-play” experience (primarily serving as a monetization opportunity for the developer, who is already a cloud model service provider).
Cloud version pricing starts from as low as $10/month via prepaid credits or an active subscription, allowing deployment of one cloud agent. Actual testing confirmed it was indeed easy to set up, though beyond the base fee, there are also monthly model subscription fees—but a free plan is available (though limited to specific free models).
From a usage perspective, the self-hosted version remains better suited for advanced users: if you already have a stable server environment, need full control over data privacy and network routing, or want to use special models not included in Portal, the self-hosted Hermes Agent is still the better choice. The cloud version, on the other hand, is ideal for individual users who want to try it out, get started quickly, and avoid dealing with infrastructure maintenance. If you’re interested in learning how to set it up, check out our previous tutorial.
Source: KOCPC Chinese