It would work the same way, you would just need to connect with your local model. For example, change the code to find the embeddings with your local model, and store that in Milvus. After that, do the inference calling your local model.

I’ve not used inference with local API, can’t help with that, but for embeddings, I used this model and it worked quite fast, plus was a top2 model in Hugging Face. Leaderboard. Model.

I didn’t do any training, just simple embed+interference.

Milvus documentation has a nice example: link. After this, you just need to use a persistent Milvus DB, instead of the ephimeral one in the documentation.

Let me know if you have further questions.

OP can also use an embedding model and work with vectorial databases for the RAG.

I use Milvus (vector DB engine; open source, can be self hosted) and OpenAI’s text-embedding-small-3 for the embedding (extreeeemely cheap). There’s also some very good open weights embed modelsln HuggingFace.

If I remember correctly, you can also use a water drop in the lens and it will amplify the image.

Back in university, I studied basically all day long, which was tiresome after long sessions of study, even if with friends. My great superpower is that it used to just take me ~10 seconds of resting with my eyes closed to feel a huuuuge boost of energy that lasted for 1-2 hours. After that boost expired, I just did it again.

Incredibly useful.

Genuine question: why would Denmark be happy/help Greenland become independent?

I manage some servers and awk can be useful to filter data. If you use commands like grep, and use the pipe operator (the " | " command), awk can be very handy.

Sure, a Python script can do that as well, but doing a one-liner in Bash is waaay faster to program.

Thanks!

If you don’t mind connecting directly with your IP address, you don’t even have to pay a domain. Been doing this for around 2 years

No direct answer here, but my tests with models from HuggingFace measured about 1.25GB of VRAM per 1B parameters.

Your GPU should be fine if you want to play around.

For the media server, I recommend taking a look at Jellifyn. If you want some fancy statistics use, also give it a look at a Prometheus+Graphana config.

Even if it’s not the case, I found the console installer to be surprisingly easy.

Based on your comment, I understand that AMD GPUs aren’t capable of 8k@60 over HDMI. Is that the case? If so, why?

The weird thing is that she seems to be an actual person, if you check the socials she links in her DM. Kind if fascinating.

Simracing with VR is quite nice, tho. You’ll never get that level of immersion with screens

The last time (some 7 months ago) something like happened it ended up being that the Steam survey had a bug that oversampled Chinesse machines. Maybe something similar happened.

He was being chased by the US government, and Assange proved that being in an US allied country will still get your arrested/tortured. What other options did Snowden had other than escaping to Russia?

IMO don’t hate the player, hate the game.

deleted by creator

Yes. Distilled models run great. You just need to chose an aproppiate version based on your hardware

Unrelated question: is that GrapheneOS?