abi

It's worse than that. It's 5-4 on the read of the constitution since Kavanaugh doesn't think it violates the 14th amendment, just a federal statute.

I think until Fable, Claude's vision was significantly worse than GPT and Gemini in my personal experience. I eval almost every vision model since I work on screenshot to code conversion project: https://github.com/abi/screenshot-to-code.

Your links are broken FYI.

https://github.com/abi/lilo I’m working on Lilo, a Telegram AI agent that can remember things, store files, track your TODOs, manage your calendar, conduct research, build apps, send you reminders and monitor things for you.

I’ve found it super useful in my personal life and is pretty much my #1 app.

What benefit would it truly provide? Companies would simply say they need to cut costs to maximize shareholder value, which is no different than what happened here.

No, we mostly spent our time on data structures and algorithms.

Ugh, good point.

Usually, those get released a few weeks later.

I'm quite confused by this article. If you persist conversation history in a database, and have all agentic turns run on the server, and merely listen to the streaming events/history via a websocket on the client, this is easily achieved. You can have as many clients as you want.

The HTTP layer is fine. Websockets work great. This is how the Codex app server works, I believe: https://openai.com/index/unlocking-the-codex-harness/ Same pattern I've used in my agentic OS/personal assistant project: https://github.com/abi/lilo Works great!

Codex is great.

Exactly.

I’ve experienced that issue as well. Clearing the cache and redownloading seemed to fix it for me. It’s an issue with the upstream library tvmjs that I need to dig deeper into. You should be totally fine on a 32gb system.

Use secret llama in a incognito window. Turn off the Internet and close the window when done.

Thanks for the bug report. Yeah, it’s a bug with not resetting the state properly when new chat is clicked. Will fix tomorrow.

Chat history shouldn’t be hard to add with local storage and Indexed DB.

Yes. Web-llm is a wrapper of tvmjs: https://github.com/apache/tvm

Just wrappers all the way down

Appreciate the kind words :)

Window AI (https://windowai.io/) is an attempt to do something like this with a browser extension.

Try to switch models to something other than tinyllama (default only because it’s the fastest to load). Mistral and Llama 3 are great.

This is amazing. Thanks both for sharing your stories. Made my day.

Well, it should be possible to just drag and drop a file/folder

Just bought the domain a couple of hours ago so DNS might not have propagated. Try back tomorrow or download and install it from GitHub (it’s just 2 steps)

Would love to. It uses MLC AIs webllm so just need to convert it to that format.

Yes, it only starts the download after you send the first message so visiting the site won’t use up any space.

Approx sizes are listed in the GitHub README.

Models are stored in indexeddb and will be managed by the browser. Might get evicted.

Is the model being released?

Our databases are all down :(

Is there a similar one for Postgres? I’d love to use it.

How? What models have they put out that are relevant?

Thanks!

There's a dropdown where you can choose a stack for screenshots (Tailwind, React, Vue, etc.). I haven't updated the prompts for the video feature just yet. You can tweak the prompt yourself here: https://github.com/abi/screenshot-to-code/blob/6069c2a118592...

The quality of the output code is solid, I think. You can see the code for the examples: https://codepen.io/abi/pen/ExJPdop and https://codepen.io/abi/pen/jORWeYB

I think the biggest thing LLM code is typically missing is better abstractions/componentization. You could probably prompt around some of that.

Thanks for clarifying. I'll send you an email once I try it.

Been looking at a few solutions similar to yours! I'm currently on Render and looking to move elsewhere so I can have more control and particularly insight into system metrics. Do you support zero downtime deploys? It wasn't clear to me from your home page.