Mamading Ceesay
Mamading Ceesay
An alternative to Electron worth considering is Tauri https://tauri.app/.
This is what the Coturn docs actually have to say on the topic: ## Performance and Load Balancing: When used as a part of an ICE solution, for VoIP connectivity,...
There's some insight on scaling Coturn here: https://webrtchacks.com/coturn-no-time-to-die-qa-with-new-project-leads/#post-3942-_gnecil8tv1mg
After a lot of pain, I solved the problem by switching to `gpt-3.5-turbo-16k`. Clearly relates to #5 and having to work around the hardcoding of GPT-4 in parts.
As noted elsewhere, I changed the hardcoding to `gpt-3.5-turbo-16k`.
I've started work on integrating https://github.com/block/goose based on their recent Docker support.
I followed up by trying a less exotic Falcon 3 quant from [tiiuae/Falcon3-1B-Instruct-GGUF](https://huggingface.co/tiiuae/Falcon3-1B-Instruct-GGUF). Running `./llamafile --verbose --chat -f watchmen_chronology.txt -m Falcon3-1B-Instruct-q4_k_m.gguf ` crashed with the following error: `llama_model_load: error loading...
I'm finding this useful for figuring out how to correctly setup `age` public keys for `agebox`. https://github.com/slok/agebox-test I now have a `.env.agebox` file.
And I was able to decrypt once I implemented `AGEBOX_PRIVATE_KEYS` as per: https://github.com/slok/agebox?tab=readme-ov-file#private-keys
I've got the encryption piece down as follows and I can manually decrypt with `sops decrypt enc.env` after setting up `$HOME/.config/sops/age/keys.txt`: #### Steps 1. Generate age keypair if you don't...