cobanov
Joined 5 years ago
Karma
61
Submissions
29
Member since
Dec 22
Recent submissions
Comment · 29 days ago
i have proxmox server at home with many vms and all of them behind cloudflare tunnel and tailscale
i also doing monitoring using self hosted ntfy notifications and when some site is slow i am connecting with ssh and checking manually, this project is really for me
i have some questions, is this agent open source, curl is ok for me but i want to see the code before
Comment · 1 month ago
Thanks!
Comment · 1 month ago
Fair, and I'd be happy if they did. Ollaya uses the same API as Jev, so your code isn't tied to it either way
Comment · 1 month ago
Depends on the model. The small ones I support today are well below Jev on harder queries, but fine for simple, well-defined questions. The open models that get close to Jev are bigger, and I'm adding support for those next.
Comment · 1 month ago
The link rgbrgb posted is a good overview. The best open ones are close to Jev now, but they're big models. And I agree, if you have an eval set for a fixed task, a trained classifier is the better choice.
Comment · 1 month ago
Fair point, that example is basically classification. I'll change it to something that looks more like a real decision.
Comment · 1 month ago
Developer here. You're right, Laya is a lot weaker than Jev, especially on harder queries. It's a small model, so it's fast, but that's the trade-off. The open models that get close to Jev are much bigger, and running those is what I'm working on next.
building ai agents with complex memory and long-running context. backend systems, vector databases, training diffusion models, and generative ai pipelines. mostly python, fastapi, and pydantic ai.
writing about ai engineering, homelabs, and tinkering on [cobanov.dev](https://cobanov.dev).
### // interests
ai agents · agent memory · backend systems · vector databases · local llm serving · diffusion models · generative art