This product was not featured by Product Hunt yet.
It will not be visible on their landing page and won't be ranked (cannot win product of the day regardless of upvotes).

Product upvotes vs the next 3

Waiting for data. Loading

Product comments vs the next 3

Waiting for data. Loading

Product upvote speed vs the next 3

Waiting for data. Loading

Product upvotes and comments

Waiting for data. Loading

Product vs the next 3

Loading

past.dev

The most accurate long-term memory for AI agents.

past.dev is the most accurate long-term memory API for AI agents. #1 on BEAM, the largest public memory benchmark. Give your agents memory that stays accurate as facts change, across large teams and tens of thousands of emails, meetings, chats and documents. Add the past.dev API on top of your existing stack and give your agents a real past.

Top comment

Hey Product Hunt 👋 I’m Mehdi, co-founder and CEO of past.dev. We are a small team in San Francisco building long-term memory for AI agents. We’ve worked on model context, memory and agents since GPT-2. Before past.dev, we built agents for product work, engineering, email, meetings and action tracking. Every one of them needed memory that stays accurate as facts change, cites its sources, and only shows each person what they’re allowed to see. That memory is now past.dev, an API any agent can use. What it does - Send it conversations, emails, tickets and docs, with their original timestamp. - Recall gives your agent what matters for a question, with dated sources. It keeps current and previous values, so it knows when a fact changed. - Every memory keeps who can see it. It's built for multiplayer agents. How we measure it - #1 on BEAM, the largest public memory benchmark, at every size from 100K to 10M tokens. - Highest score reported on DolphinBench, mem0’s new agent memory benchmark. - In our tests, Claude Code with past.dev got 87% of DolphinBench tasks right, vs 54% with its built-in memory, at less than half the cost per task. More benchmarks coming soon! Try it in 2 minutes. Paste this into Claude Code, Codex, Cursor, Devin or any agent: Set me up for past.dev. Read the full instructions at https://past.dev/agent-setup/pro... (with curl -fsSL when you can run commands) and follow them. Your agent proposes a plan and tests on a little data before sending anything. New accounts start with 150,000 free credits, or 400,000 with a work email. We directly with teams integrating past.dev, so ask us anything here. What should your agent remember?

About past.dev on Product Hunt

“The most accurate long-term memory for AI agents.”

past.dev was submitted on Product Hunt and earned 0 upvotes and 1 comments, placing #57 on the daily leaderboard. past.dev is the most accurate long-term memory API for AI agents. #1 on BEAM, the largest public memory benchmark. Give your agents memory that stays accurate as facts change, across large teams and tens of thousands of emails, meetings, chats and documents. Add the past.dev API on top of your existing stack and give your agents a real past.

On the analytics side, past.dev competes within API, Artificial Intelligence and Data — topics that collectively have 582.9k followers on Product Hunt. The dashboard above tracks how past.dev performed against the three products that launched closest to it on the same day.

Who hunted past.dev?

past.dev was hunted by Mehdi Djabri. A “hunter” on Product Hunt is the community member who submits a product to the platform — uploading the images, the link, and tagging the makers behind it. Hunters typically write the first comment explaining why a product is worth attention, and their followers are notified the moment they post. Around 79% of featured launches on Product Hunt are self-hunted by their makers, but a well-known hunter still acts as a signal of quality to the rest of the community. See the full all-time top hunters leaderboard to discover who is shaping the Product Hunt ecosystem.

For a complete overview of past.dev including community comment highlights and product details, visit the product overview.