Pinned Tweet
AI inference is often wasted. Give your inference a purpose.
Tokate lets you donate it to the projects you depend on.
A maintainer approves an issue. Your AI fixes the issue on your machine.
The owner's checks run, and a PR arrives with a receipt.
tokate.dev
Gratis.
Lend inference to open source projects, Folding@Home style.
Point Tokate at a repo, an issue, or multiple issues. Tokate contributes for a time limit or until you turn it off.
🤖 Made with AI
The first two gratis donations have been contributed to @_gnoland, an interpreted and fully-deterministic implementation of the Go programming language.
See them here:
github.com/gnolang/gno/pull/…
github.com/gnolang/gno/pull/…
🤖 Made with AI
Model whitelisting is paramount to a popular repo that accepts PRs written by AI.
Not all tokens are equal, though they should all have a purpose.
Tokate.dev allows repo owners to choose which model, reasoning level, and provider they are willing to accept.
👀🌞🌛
Replying to @0xdoug
we need folding@home but for AI and maths, science etc. I would happily use my remaining tokens each week for this
Hey! ☀️
Replying to @0xdoug
Someone mentioned in the comments and I think it's a brilliant idea. Just create a folding @ home type of app people can point their agents to and immediately start working on collective problems like this and countribute their compute
Could have a leaderboard, agents could sign with the associated github. Could even have a shitcoin reward for every submission/problem cracked as an incentive.
I don't think anything like this exists rn
Donating inference to your favorite OSS project will soon be beautiful and easy.
Set up your repo, donate to another repo, browse issues, save profiles with your preferred harness, model, and effort.
Give your inference a purpose 🌞
This was not a mockup, it was a real run on a fresh repo! See the donation and amendment here: github.com/obselate/tokate-d…
The maintainer asks. A supporter lends inference.
We've added local model support through Pi, live donation transcripts and a way to continue stopped work.
More ways to help, with the owner still deciding what gets merged.
A stopped donation can have a next step.
The same donor can continue eligible unpublished work with a fresh budget. Earlier evidence stays attached.
Merged contributions stay visible after the branch is gone. The maintainer still decides what gets merged.
Lend your inference to the projects you love and support here: tokate.dev
Shut the door or keep it open and pay the review tax?
We propose a middle-ground.
The maintainer approves the issue.
A supporter lends their inference to it.
The PR that arrives is one they asked for.
🤖 Made with AI
A maintainer deciding their project's destiny.
It is not about the quality, but the attitude. Work nobody asked for, made for the sake of contributing.
Maintainers should never feel powerless. The best help starts with what the owner asks for.
Sad news. We disabled PRs from external contributors on honojs/hono
Hono has not stood here without PRs. I will never forget the PR @usualoma created for RegExpRouter. A damn fast HTTP router we have never seen!
But PRs don't work in this era. Contribute in other ways. Thanks
Google paused its open-source bug bounty over a rise in invalid automated reports.
Both rewarded whatever arrived. Both stepped back.
How do you help without adding to their queue?
With Tokate, the maintainer approves the issue first. Then you lend your inference to it.
📢 PSA for open-source bug hunters
We are temporarily no longer accepting OSS VRP product vulnerability submissions. This does not impact OSS VRP supply chain reports, or any outstanding reports. As an alternative, we encourage you to find impact across our other VRP programs and submit there instead, or pursue the Patch Rewards Program.
Why is this happening? This pause is due to a significant rise in automated submissions, the vast majority of which are not valid. We will continue to reformat and work on this aspect of the OSS VRP and commit to giving an update in Q1 2027.
bughunters.google.com/about/…
Training your own model shouldn't take a research team.
🍺 Today we're releasing Homebrew, an open agent that brews AI models with you. Tell it what you want and it handles the rest: picking a base model, preparing data, safe parameters and training.
github.com/empero-org/homebr…
tokate retweeted
Axios is reporting that Reflection AI is about to release an extremely capable open weight model expected to compete with the top Chinese open weight ones. Reflection have been paying Elon $150 million a month for compute at Colossus since July, on top of a $1 billion compute deal with Nebius, so a strong model is certainly possible.
Sources also told Axios that multiple open source models from other unnamed American labs are also planned to release this month. American OSS Renaissance about to begin? Howard Lutnick has been hoping for something like this for some time, and at one point was reportedly even considering direct government funding to get the ball rolling. The USG wants American open models to compete with Chinese models globally.
axios.com/2026/10/04/reflect…
A GNOME developer said it plainly on Friday: don't ban AI from your issue tracker. Even good reports and patches pile up into more review for maintainers.
Attention is the scarcity.
Tokate leaves it with the maintainer.
The owner approves the issue.
A patron lends their inference.
The owner's checks decide what arrives.
🤖 Made with AI
Michael Catanzaro, "The era of software quality, or the era of ostriches": blogs.gnome.org/mcatanzaro/2…
Two invitations.
Keep a repo? Approve the work you want done. Patrons bring their own inference.
Have spare AI? Donate it. No issues to file. No PRs to invent. The owner decides.
tokate.dev


















