AeionClaw Pricing — Free, No Tiers, BYO LLM
AeionClaw is included free with every Aeion deployment — the full engine, not a stripped-down trial. Unlimited personas, skills, reactions, schedules, and knowledge base entries. You bring your own AI provider key and run on your own tenant infrastructure, so there's nothing for Aeion to meter or upcharge. Real-time per-tenant spend caps keep your own AI provider bill predictable.
What You Actually Pay — Worked Example
AeionClaw doesn't charge per token, per message, or a platform subscription. You pay your chosen AI provider directly. Here's a typical deployment running 5 hours/day across a 10-person team:
Per-Tenant Cost Caps — Prevent Billing Surprises
Every deployment supports hard cost caps that prevent runaway AI spend, on daily, weekly, or monthly periods — no tier requirement.
Provider Cost Cheat Sheet
OpenAI GPT-4o
$5 in / $15 out per 1M tokens
OpenAI GPT-4o-mini
$0.15 in / $0.60 out per 1M
Anthropic Claude 3.7 Sonnet
$3 in / $15 out per 1M
Anthropic Claude Haiku
$0.80 in / $4 out per 1M
Google Gemini 2.5 Flash
$0.10 in / $0.40 out per 1M
Google Gemini 2.5 Pro
$1.25 in / $5 out per 1M
DeepSeek V4
$0.55 in / $2.19 out per 1M (reasoning)
Groq (Llama 3.1)
$0.30 in / $0.60 out per 1M (ultra-fast)
Ollama (local or Cloud)
local is $0 + your GPU electricity; Ollama Cloud is metered by Ollama directly
Pricing FAQ
No. AeionClaw is one flat, fully-featured experience on every Aeion deployment — no Pro or Enterprise upgrade gates any persona, skill, reaction, schedule, knowledge base, or workflow limit. It runs on your own tenant infrastructure with your own AI provider key, so there's no per-feature cost for Aeion to recoup.
ZeroClaw is how you talk to and operate the rest of the OS — personas, skills, and workflows are the harness for Aeion's MCP and WebMCP tools. Keeping it free removes any friction between "using Aeion" and "using AI to run Aeion."
Aeion's cost caps prevent that. Set your daily, weekly, or monthly USD cap; AeionClaw enforces it before each call. You'll never get a surprise bill from your AI provider. The cap is conservative by design — overage prevention, not throttling.
No — Aeion doesn't fine-tune anything. Your data stays in your tenant. If you want a fine-tuned model, do it directly with the provider (OpenAI, Anthropic) and bring the fine-tuned model ID via BYOM.
No. Add as many users as you want; cost is gated by your own AI usage volume, not headcount or Aeion feature tiers.
AeionClaw runs against your tenant's own infrastructure and AI provider — including local models via Ollama, vLLM, or HuggingFace TGI. Talk to us about your specific deployment topology if you have air-gapped or fully offline requirements.
Your data stays in your tenant database under your control. Use Aegis to take a final snapshot and export. AeionClaw doesn't hold data in any external location — nothing to migrate out.