LibreAGI functions as a private BitTorrent tracker that archives open-weights releases strictly in safetensors and GGUF formats under Apache-2.0, MIT, or Qwen Community licenses. Human administrators verify all submissions against original source commits to prevent tampering and guarantee authenticity. Participants earn system credits based on size, duration, and file rarity, including a 4x multiplier for models possessing only one surviving seeder.
Key points
A signed public catalog detailing file hashes and sources allows independent swarm recove…: A signed public catalog detailing file hashes and sources allows independent swarm recovery if the tracker fails.
The network relies on an incentive model that scales user rewards by file size, hosting d…: The network relies on an incentive model that scales user rewards by file size, hosting duration, and rarity.
Prospective participants can contact administrators using TLS via port 6697 on the projec…: Prospective participants can contact administrators using TLS via port 6697 on the project's IRC server.
Sources checked by our analysis pipeline: 1
Seeding pays by scarcity
Credit grows with size and time, and a model with one copy earns four times the base rate. The rare ones need the most help.
Open weights, members only
Open models should outlive the places that host them. LibreAGI is a private BitTorrent tracker where members keep copies of the models most likely to disappear.
$ mktorrent -p -s libreagi -l 24 ./model
Credit grows with size and time, and a model with one copy earns four times the base rate. The rare ones need the most help.
Only safetensors and GGUF, only under Apache-2.0, MIT or the Qwen Community License. Each file is checked against its source commit, and staff review every model.
A signed public catalog lists every model, hash and source, so anyone can rebuild the swarms if this site goes away.
The public channel is open to anyone, with no account needed. Come and ask questions before you apply.
An unauthorized intrusion into an internal cloud development system at Fireworks AI exposed client records alongside conversational data. The breach primarily targeted large language model compute capacity, allowing attackers to monetize unauthorized access tokens on external markets. The incident underscores growing supply-chain vulnerabilities for enterprise software platforms that integrate third-party artificial intelligence infrastructure.
Hardware makers are entering the wearable artificial intelligence market by introducing advanced frames integrated with Google's Gemini technology. These new connected devices are positioned to reach everyday consumers through an entry-level price point of $395. The launch marks a critical step in expanding ambient computing tools directly into user eyewear.
Federal healthcare initiatives face heightened scrutiny following the disclosure of fresh records detailing algorithmic review mechanisms in public coverage. The revelations come at a pivotal moment as sector leaders and policymakers gather for major industry summits such as HLTH. Understanding the full implications of these findings remains limited due to restricted reporting access.