Replies: 1 comment 1 reply
-
The project is shit, the reason is no rigid types, the second aspect, 3300 types of neurons. And what do you have in your project,??? Signal attenuation? solved the problem.
|
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment

Uh oh!
There was an error while loading. Please reload this page.
-
Hi — wanted to share what we built using Odysseus as the AI backend.
Synaphos (https://synaphos.com) is a self-hosted web dashboard that connects directly to Ollama. We added an on-chain layer: every AI response can be attested on Base L2 via EAS and stored on IPFS via Pinata.
Running it in production on a VPS with llama3.2.
A few things we ran into during setup that might be useful for the docs:
Source: https://github.com/synaphos/synaphos
Open to contributing doc fixes or anything else useful to the project.
Beta Was this translation helpful? Give feedback.
All reactions