📬 Email for AI agents, not humans!
– No CAPTCHA, No human signup
– Register via API in ~30 seconds
– Connect via MCP or Agent Skill
– Runs the whole workflow
Free in open alpha - link in comments
@danyrayleanu@AgnoHQ Hi! We apologize for the problems you are having. We are already fixing everything and everything will be stable by Tuesday. Heavy load on servers
Give your Agno agents their own inbox with Atomic Mail!
@AgnoHQ lets you build AI agents, now they can create an inbox in 30 seconds. It send, read, search and reply to emails - all by themselves.
Watch how our support agent does it.
Connect your agent to its own email
MiniMax M3 beat Claude Opus 5 58x on cost
MiniMax-M3: 31K tokens, $0.018
Gemini 3.7 Flash: 25.6K tokens, $0.021
GPT-5.6: 29.1K tokens, $0.128
Claude Opus 5: 24K tokens, $1.056
M3 won on cost by miles – one clean pass. Opus 5 needed four tool calls, re-checking the same reference file each time before it sent anything. GPT-5.6 mis-parsed its own output once and redid it. Gemini took two passes.
Multiply it by a few hundred runs and the gap stops being pennies: $9 vs $528 for the same result.
Give your AI agent its own email – Atomic Mail Agentic
Love seeing real-world agent benchmarks like this 👇
MiniMax-M3: $0.018 to spin up an inbox, write and send a business email end-to-end. The cheapest model on the board - and it finished the job.
This is what efficient intelligence looks like.
(Yes! We’re a frontier lab - we don't just build video and music models, we also train and open-source our frontier LLMs for you to use.)
@MiniMax_AI MiniMax, thanks for the support! M3 + an inbox built for agents (send, wait, reply, repeat) is a genuinely cheap way to run that loop. We built the email infra for exactly this – https://t.co/BSN7SvI9jy
Your agent can now sign up for itself
It registers, grabs its own OTP from inbox and finishes the flow solo.
Try it yourself: connect Atomic Mail via MCP or Agent Skill and let it register somewhere without you.
@GhiffariZaid@0xfrsmln We think that your bills is your responsibility, not ours.
Do your research of bug bounty programs their duration and conditions before investing your time.
Anyways, we will re-check your submission with a board and come back.
@GhiffariZaid 2/2 Please don’t mislead people with SCAM accusations. We had a bounty in a very beginning and security whitepaper isn’t wrong in this term.
But now there is no active BB.
Anyways we will assess your report and come back.
Our support manager was wrong counting this as a mistake.
1/2 Hello Zheev.
For first: Thank you for dedicating your time.
We will re-check your report on a board level.
Second: Send us an active atomic mail bug bounty program with declared rewards and other conditions.
Spoiler: there is NO active bug bounty programs.
Please don’t mislead people.
Agent Zero beat Hermes Agent 3x on tokens
We gave both agent frameworks the same prompt:
1. Register a new inbox Atomic Mail
2. Send an email
Both running Claude Fable 5
Results:
Agent Zero: 18.7K tokens / 55s
Hermes Agent: 54K tokens / 1m 05s
Both agents finished the job without mistakes. No wrong steps, no retries. But they got there in different ways. @NousResearch remembers things across chats and has a lot of tools ready to use. That takes extra tokens, even for a small job like this one. @Agent0ai keeps things light. It sends less text to the model, so it finishes faster and cheaper. If you want an agent that remembers you next time, that extra cost is worth it. If you just need an inbox made and email sent, Agent Zero is the cheaper way to do it.
Give your AI agent its own email - Atomic Mail