Using our LEM Runtime, a model that fits on your phone, wrote this over 10 turns, no history reply, just the next turn.
Model: https://t.co/0EHoZMbt4y a Gemma4-E2B-IT retune.
New revision, our first was using day-1 forks.
https://t.co/0EHoZMbt4y
ollama run https://t.co/P57bf9ze8q
Apple Native: mlx_lm.server --model "lthn/lemer"
LM Studio, just click below with it open.
We want to introduce, Lemer.
The first of 4 LetheanEthicalModel (LEM) models for Gemma4.
Adapters to re-introduce the ethics speciality in our Gemma3 models to come later.
Compatable with EU AI laws, EUPL-1.2, requires around 4GB mem 9.5 for bf16.
https://t.co/0EHoZMbt4y
We want to introduce, Lemer.
The first of 4 LetheanEthicalModel (LEM) models for Gemma4.
Adapters to re-introduce the ethics speciality in our Gemma3 models to come later.
Compatable with EU AI laws, EUPL-1.2, requires around 4GB mem 9.5 for bf16.
https://t.co/0EHoZMbt4y
@M17361358 Still active, the `iz` chain is not, the new chain is up, just making some adjustments to it for network use and the swap will open.
You'll send a tx from your `iz` wallet to `iThn` wallet, all on chain, no exchanges needed.
To little fan fair today, we would like to introduce.
Lethean Ethical Modal
https://t.co/vRlhbIvudw
The LEK Method: Ethical Kernel Fine-Tuning as an Alternative to RLHF Behavioural Conditioning: https://t.co/EdB1a5rzmW
Also, evendece Unsenored is dangeours & drops quality.
@c1sc0r1b31r0 The community team iterated quite a bit with R&D.
Keep an eye out for when we talk about https://t.co/6eTE7Uzmy9 because the distributed network would be somewhat deployed, just not opened to other gateway services.
The next few posts would be worth looking into.
Welcome to a better Google search; literally.
Why not try an AI from the EU, guarded by strict privacy laws?
Your questions are not used for training, period, end of.
(not even an affiliate-linked promo, it's just good, I think you should try it)
https://t.co/0oSesXm4z7