Launching today
Forgestack
Open-source LLM stacks, deployed in your cloud in minutes
1 follower
Open-source LLM stacks, deployed in your cloud in minutes
1 follower
Pick one of 20 open models, connect AWS or GCP, and ForgeStack deploys the model, a RAG knowledge base and a chat API inside your own account. Your data never leaves your cloud. Fine-tune with LoRA, swap models without changing the API. Free during launch.
Why you built it: for example, teams that want an LLM over internal docs but can't send that data to OpenAI.
How it works, in three lines.
Honest costs: the free-tier models run on t3.micro, the 7B models need a GPU at about $1/hr, and the cloud provider bills users directly.
One specific question to start discussion, for example: "Which models should we add next?" or "AWS or GCP first for you?"