aichannel

Introducing a smol Hermes 3 LLM!

Hermes 3 3B is now available on @huggingface alongside quantized GGUF versions to make it even smaller.

More info and download links here: https://t.co/WIVBuWK2Sf

Run Hermes on phones, laptops, and CPUs without sacrificing speed, and may also… https://t.co/ad8sjG7eb1

Nous Research

Can someone explain why this is useful?

https://x.com/NousResearch/status/1866920983187292290

There is a huge community focused on local llm use. A smaller subset likes the idea of a local LLM on phones. I've seen this in various discussions within the community and while this is a small model that will not perform well, it will meet this communities ethos of local first. It's useful to them, though not objectively useful for anything else.

on agents stuff small models are useful for some of the smaller sub-tasks or for routing through all the optional tools (and cheaper)

all of these features are hype

Hermes 3 contains advanced long-term context retention and multi-turn conversation capability, complex roleplaying and internal monologue abilities, and enhanced agentic function-calling.