Avatar
pleblee
69a0a0910b49a1dbfbc4e4f10df22b5806af5403a228267638f2e908c968228d
You should not need an account on a corporate server in order to use software.

Running keychat

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

nostr:nevent1qqsvy5qvjpj86dd8me3hrwhyzadakehvgjpvraps33xdtjc9lgax50gpzamhxue69uhky6t5vdhkjmn9wgh8xmmrd9skctczyzaljga2jfrqvhugcsx8mxlkrnxvplelelcxt2xt9l6vlwmzpz83uqcyqqqqqqgqs5t6n

Yeah, I like'em but they don't update their models very much

Oh and I autostart it on boot with a systemd unit in my `configuration.nix`:

```

systemd = {

services.ollama-webui = {

wantedBy = [ "multi-user.target" ];

after = [ "network.target" ];

serviceConfig = {

Type = "forking";

User = "pleblee";

WorkingDirectory = "/home/blee/apps/open-webui";

Environment = "NIX_PATH=nixpkgs=/nix/var/nix/profiles/per-user/root/channels/nixos:nixos-config=/etc/nixos/configuration.nix:/nix/var/nix/profiles/per-user/root/channels";

};

script = "${pkgs.nix}/bin/nix-shell";

};

};

```

local llms with ollama + open-webui + litellm configured with some apis

the free Claude 3.5 sonnet is my daily driver now. When I run out of free messages, I switch over to open-webui and have access to various flagship models for a few cents. The local models usually suffice when I'm asking a dumb question

Here's my `shell.nix`:

```

{ pkgs ? import {} }:

(pkgs.buildFHSEnv {

name = "simple-fhs-env";

targetPkgs = pkgs: with pkgs; [

tmux

bash

python311

];

runScript = ''

#!/usr/bin/env bash

set -x

set -e

source .venv/bin/activate

tmux new-session -d -s textgen

tmux send-keys -t textgen "open-webui serve" C-m

tmux split-window -v -t textgen

tmux send-keys -t textgen "LITELLM_MASTER_KEY=hunter2 litellm --config litellm.yaml --port 8031" C-m

tmux attach -t textgen

'';

}).env

```

And here's `litellm.yaml`:

```

model_list:

- model_name: codestral

litellm_params:

model: mistral/codestral-latest

api_key: hunter2

- model_name: claude-3.5

litellm_params:

model: anthropic/claude-3-5-sonnet-20240620

api_key: sk-hunter2

- model_name: gemini-pro

litellm_params:

model: gemini/gemini-1.5-pro-latest

api_key: hunter2

safety_settings:

- category: HARM_CATEGORY_HARASSMENT

threshold: BLOCK_NONE

- category: HARM_CATEGORY_HATE_SPEECH

threshold: BLOCK_NONE

- category: HARM_CATEGORY_SEXUALLY_EXPLICIT

threshold: BLOCK_NONE

- category: HARM_CATEGORY_DANGEROUS_CONTENT

threshold: BLOCK_NONE

```

Replying to Avatar pleblee

https://bitcoiner.social/static/attachments/4rG1Kg_tony--d6_trim.mp4

I had way too much with LivePortrait this morning. Couldn't quite get it to run on Nix (I still suck at nix derivations), but they're still running a free demo space on HF.

Also out of China in the last few weeks was ChatTTS and Qwen2, state of the art local text to speech and llm. I've been running it with ollama and open-webui, and it's not half bad compared to 4o or claude, in one prompt it actually gave me a better answer.

There's a ComfyUI node for LivePortrait. There's a lot you could do with this stuff. Deepfakes and destroy democracy, sure. But maybe more relevant to my interests is we're nearly upon the toolset needed to build a fully artificial (and local) talking head. if there's a model that can generate facial expressions to match a voice, I think that's basically everything we need is right there. then use ollama to RAG your own documents and legit talk to your home PC face to face.

what a time to be alive

Replying to Avatar pleblee

https://bitcoiner.social/static/attachments/4rG1Kg_tony--d6_trim.mp4

I had way too much with LivePortrait this morning. Couldn't quite get it to run on Nix (I still suck at nix derivations), but they're still running a free demo space on HF.

Also out of China in the last few weeks was ChatTTS and Qwen2, state of the art local text to speech and llm. I've been running it with ollama and open-webui, and it's not half bad compared to 4o or claude, in one prompt it actually gave me a better answer.

There's a ComfyUI node for LivePortrait. There's a lot you could do with this stuff. Deepfakes and destroy democracy, sure. But maybe more relevant to my interests is we're nearly upon the toolset needed to build a fully artificial (and local) talking head. if there's a model that can generate facial expressions to match a voice, I think that's basically everything we need is right there. then use ollama to RAG your own documents and legit talk to your home PC face to face.

what a time to be alive

state of the art open weights*

https://bitcoiner.social/static/attachments/4rG1Kg_tony--d6_trim.mp4

I had way too much with LivePortrait this morning. Couldn't quite get it to run on Nix (I still suck at nix derivations), but they're still running a free demo space on HF.

Also out of China in the last few weeks was ChatTTS and Qwen2, state of the art local text to speech and llm. I've been running it with ollama and open-webui, and it's not half bad compared to 4o or claude, in one prompt it actually gave me a better answer.

There's a ComfyUI node for LivePortrait. There's a lot you could do with this stuff. Deepfakes and destroy democracy, sure. But maybe more relevant to my interests is we're nearly upon the toolset needed to build a fully artificial (and local) talking head. if there's a model that can generate facial expressions to match a voice, I think that's basically everything we need is right there. then use ollama to RAG your own documents and legit talk to your home PC face to face.

what a time to be alive

oh I posted an animated video based on your profile image

Replying to Avatar calle

This article by nostr:npub1l6uy9chxyn943cmylrmukd3uqdq8h623nt2gxfh4rruhdv64zpvsx6zvtg gives an excellent introduction to Ecash:

Chaumian Ecash implementations backed by Bitcoin are kicking up a lot of developer activity today, and for good reason: By reviving a well-studied technology we can offer ordinary people a new wave of usability improvements, making Bitcoin-powered payment systems more approachable to businesses and customers alike.

An Ecash mint issues Ecash notes in various denominations, which can be redeemed later at the mint for some fungible commodity or service (such as Bitcoin). This is conceptually very similar to the physical banks of ye olden days, which accepted deposits of some fungible asset like silver or gold, and issued physical cash (paper notes or metal coins) of various denominations in return, which could be used to reclaim the equivalent amount of the physical asset.

In the case of a Bitcoin-backed Ecash mint, the mint accepts Bitcoin deposits, and issues Ecash notes which can be redeemed for Bitcoin at a future time, for as long as the mint remains solvent and operating.

The basic principles of an Ecash mint are:

- Authenticity - The Ecash mint can be certain the Ecash notes it creates cannot be forged by others, so that when a depositor redeems a note, the mint is safe against fraudulent withdrawals.

- Fungibility - Depositors can be certain the Ecash notes they receive are anonymous and fungible. Instead of the mint giving a depositor a bearer token (which would have to be recognizable by the mint later when redeemed), the depositor cooperates with the mint to blind the Ecash notes they receive through a clever cryptographic protocol called blind signatures.

https://conduition.io/cryptography/ecash-dlc/

I have been wanting to study this stuff recently, thank you.

A lot of other stuff breaks when doing that currently. I tried creating a new npub recently and couldn't load a community that I created on this npub.