Відео

Awesome video. I’m from America and I visited Wales and England earlier this year for the first time. Wales was so beautiful.

What is the pc or laptop min config needed to run ollama llama 3 model and mixtral model, pls let us know

Your mouth is way to wet

Amerikanische Verhältnisse light. Ekelhaft!

Initially PostgREST appears awesome. But then we see that the documentation is lacking and also missing a real world example repository. 1. I was not able to find a *recent* example repository with lots of migrations, views, functions. In GitHub discussions, the suggestion from owner for migrations was to package as PG EXTENSION. But again no example repository. 2. An example of how devops will look like is missing. 3. If the developer also needs to access DB, now [s]he needs ORM code which has to kept in sync with table schema created from migrations (sqitch sql scripts for row level security with policies). Or is it like the developer should always use the postgrest http client and never touch DB directly.

Aider is pretty interesting I was trying it out using Anthropic. Did the job for me and I want to play more with it.

Instructor patched SDK worked good as well, but this just cleans up the code and is simple. Based on my own implementation with instructor, it obeys the Pydantic basemodel structures close to a 100% of the completions, compared to the few-shot propmting setup i had before that, that had about 10% success and LOADS of failures 🙂 This way just works !

But given that OpenAI, Claude can do this already, why do we need ollama for it? And here, isn’t Llama doing the heavy lifting? Not clear how ollama is helping.

I’d appreciate if you could share the codes.

👋👋👋

Ive been needing this feature, thanks for the info!

Great adavancment for open source efforts

Interesting, is it exposing swagger ui like ?

Awesome, one more tool to master. Wasn't enough with pip, pipx, pipbla blatry etc. Super. The great thing about standards is there are so many to choose from!

Thanks Ian, I'd looked at and liked the look of postgrest. Your video sells it to me !

can it be used for kyc verifications ?

getting erro "'uv' is not recognized as an internal or external command, operable program or batch file. " but I have already installed uv. I am using windows

Is it preferred to move all packages from requirements.txt to pyproject.toml?

Can Ollama benefit from an NPU hat? The Hailo 26 TOPS for instance.

what if you are already using pyenv, how to replace with uv?

and they made fun of npm and yarn...

This is LLMs running on CPU... yikes

dullest man on UA-cam... zzzzzz

Ollama2 no longer runs on my raspberry pi5 8G, there is error message that models requiring system memory (8.4G) than is availabel(7.4G), can it be resolved or anything I got wrong?

can i use this on a 2015 Mac

how can I use my own client? as many company has their own client originate from openai. I try to use it, but it tell me absent of provider

Is this version of face fusion software unfiltered and unleashed or does it have nsfw protection filters? John E

Streamlit is better.

hello , I 'm descovering fastHTML and the HTMX scope , and I'm wandering about the capabilities of this framework especially if i want to make a heavenly front app ( because of some webmapping that needs some frameworks and maybe javascript intervention ) ? if i have a template can i convert it to component of fasthtml very easily ?

your keystrokes sound like drum taps.

Thank you for making this video. I've always wondered how these LLMs perform on low-end devices. Obviously, not great, but this looks promising!

kinda seems like a redundent question since its based on arch. but you could dual boot any os.

It's actually unneccessary to use apsw - you can just use sqlite3 in Pythons standard lib. Adding the function takes an extra argument, but I've added a couple of examples to the repo.

How do you do a rootless docker multistage build using uv? Preferably zero CVE with something like chain guard wolfi. Can't get it to work.

Schaden Ways

Something like rye, isn't it? For python I prefer things written in python.

Hello, don’t work more, can U update please? ❤

how do we stream?

Why not use DSPy?

Interesting, I’ve been building out relational neural networks that use structured data for enhanced query efficiency. Keen to see how it challenges LMMs

thanks for introducing a new tool for my work! Subscribed <3

Ths seems like an amazing tool for building the prompts and debugging but I can't see this being used in any large project in production. You need to have full control over the API calls and often optimize on that level.

wow that's even faster than my pc idk how

soy devs are gonna love this

How is this not known out there??? Great find.

Decided to give it a try in my current project instead of LangChain. So far, I'm impressed. Let's see when it comes to RAG.

I would have understood a simplification by returning a tuple, where it can be just two strings, with the first one being system and the second user, by default, overridable by decorating it with ell.system etc. for complex situations, but using the docstring, idk if I find that clean, otherwise very interesting toolset

prompt engineer joke. Engineer, sure! More like prompt monkey.

wonderful intro of ell, thanks. I am trying it with claude: @ell.simple(model="claude-3-5-sonnet-20240620"), does it automatically know that the provider is switched from openai to anthropic？

Great video Ian. I've only just discovered FastHTML and was looking for more information about it, and there you were!