Hermes Agent vs vast.ai
Pricing, plans and key features compared side by side. Updated November 2026.
Pricing model
Freemium
Paid
Free plan
✓ Yes
✕ No
Paid plans from
$20/month
From $0.03/hr
Who it's for
Hermes stands out because it is genuinely self-improving -- the soul.md and context file system means the agent builds up knowledge about your specific workflows over time rather than starting fresh every session. The combination of self-hosting (so your data stays yours), open-source extensibility, and a growing skills ecosystem makes it the most customizable general-purpose agent available. If you are comfortable running software on your own machine, it is the easiest way to get a persistent AI agent that actually knows you.
vast.ai is the go-to GPU marketplace for AI builders who need serious compute without enterprise pricing. Whether you are fine-tuning a model, running batch inference, hosting a Stable Diffusion endpoint, or spinning up a local LLM server, vast.ai gives you access to a global pool of GPUs at dramatically lower prices than the major clouds. The platform is flexible, transparent on pricing, and simple enough to spin up in minutes.
Key features
- Multi-Platform Connectivity -- Connect Hermes to Telegram, Discord, Slack, WhatsApp, Signal, Email, and CLI so the agent meets you wherever you work
- Persistent Memory -- Hermes learns from your projects via soul.md, agents.md, and context files; preferences and solutions carry forward across every session
- Cron Scheduler -- Schedule recurring tasks in plain language -- reports, backups, monitoring jobs -- without writing cron syntax or managing separate automation tools
- Delegated Subagents -- Spawn isolated sub-agents with their own conversation history, execution environment, and tool access for parallel or sensitive workloads
- 300+ Models via Nous Portal -- Access OpenAI, Anthropic, Google, Mistral, and Nous own models through one subscription with bundled web search, browser automation, image gen, and TTS
- Five Execution Backends -- Run tasks locally, inside Docker, over SSH, in Singularity containers, or on Modal cloud; swap backends per task without changing your agent setup
- Skills System -- Package repeatable workflows as named skills (e.g. /review-pr, /deploy-staging) that you or your team can invoke with a single command
- Hermes Desktop -- Native cross-platform app (Mac, Windows, Linux) with a streaming chat interface, file browser, preview rail, config UI, and multi-agent orchestration panel
- GPU Marketplace rent H100, A100, RTX 4090, 5090 and hundreds of other GPUs from providers worldwide
- 10–20x cheaper than AWS or GCP GPU instances pay only for what you use
- On-demand and interruptible instance types to match your budget and workload
- Docker-based environments bring your own image or start from popular ML templates
- SSH access and Jupyter notebooks included out of the box
- Global provider network filter by location, CUDA version, GPU RAM, and price
All plans
Free (Self-Hosted)
$0
Full Hermes Agent with your own API keys; MIT-licensed, no usage limits imposed by Nous, unlimited installs
Plus
$20/month
$22 in Nous Portal credits per month; access to 300+ models, web search, browser automation, image generation, and TTS bundled
Super
$100/month
$110 in Nous Portal credits per month; same bundle as Plus at higher volume; suited for daily heavy use
Ultra
$200/month
$220 in Nous Portal credits per month; highest volume tier for power users and teams running continuous agent workloads
Interruptible
From $0.03/hr
Cheapest option — great for fault-tolerant training jobs that can checkpoint and resume
On-demand
From $0.10/hr
Reserved access — ideal for inference endpoints and interactive workloads
Pay-as-you-go
Varies by GPU
No subscription required — rent by the hour across hundreds of GPU configurations
Profile views
262
on OnePlaceForAI
240
on OnePlaceForAI
Add another agentic workflow tool
New AI tools, every Monday
Free weekly digest — tool picks, prompts, and AI news.
+ Token Board