llmllmllm

The UK doesn't have exit checks.

I just made this after realising that chatgpt 4 now has 32k input context so it's useful to inject multiple files into my prompts easily

"Show HN is for something you've made that other people can play with. HN users can try it out, give you feedback, and ask questions in the thread."

We made it, you can play with it :)

This is an interface that has GPT-4 producing content while also producing images using DALL-E 3, we have integrated them.

Feedback is great, thank you.

I linked to the shared report so that it's obvious what is actually being produced, rather than just a description of it.

This is just an example summary of a single web page, you could imagine it producing much more compelling content, for example combining multiple pieces of information, along with a twist, just as many humans do when publishing on the web. FlowChai could search hundreds of documents to produce a report (using RAG), so in that sense it can go beyond what a person could do in a reasonable time frame.

While this makes some of what my startup https://flowch.ai does a commodity (file uploads and embeddings based queries are an example, but we'll see how well they do it - chunking and querying with RAG isn't easy to do well), the lower prices of models make my overall platform way better value, so I'd say overall it's a big positive.

Speaking more generally, there's always room for multiple players, especially in specific niches.

I have found success with chunking with 100 tokens, preceeded by the last 10 tokens of the previous chunk, and the first 10 tokens of the next chunk, 120 tokens total. I generate an embedding for each, then compare that to embedding(s) derived from the input query.

How to generate embeddings from the input query well is where one's focus should be IMO. An example: "don't mention x" being turned into filtering out / de-emphasizing chunks that align with the embedding for x.

I've been using these techniques along with pgvector and OpenAI's embeddings for https://flowch.ai and it works really well. A user uploads a document or uses the Chrome Extension on a webpage and FlowChai chunks up the content, generates embeddings, builds up a RAG context and then produces a report based on the user's prompt.

I hope that helps show a real world example. You're welcome to play with FlowChai for free to see how it works in practice at the application level.

Things like this needing to exist are one of the reasons we are working on https://flowch.ai/?ref=hn - the ChatGPT interface is really sub-optimal for work and keeping track of interactions over time. We have been building a much better way to organize LLM interactions, even basics like splitting things into projects and search make it a much better experience.

I'll be curious to see if many people end up subscribed to both this and ChatGPT Pro, are they different enough to both be worth it?

Very fair, we have demo videos, guides etc planned for the next week or so. As it's a tool that can do many things it's hard to describe. Still a lot to do :)

In terms of what makes the report different from Bing: this could be any source of data: scraped from the web, search, API upload, file upload etc, so there's a lot more power there. Also, it's not just one off reports, there's automation there which would allow for example a weekly report on the latest papers on GPT4 (or whatever you're interested in).

Interesting that they're still centered around Chat as the interface, with https://flowch.ai (our product) we're building it much more around projects and reports, which we think is often more suitable for businesses.

We're going after some of these use cases:

Want a daily email with all the latest news from your custom data source (or Google) for a topic? How about parsing custom data and scores from your datasets using prompts with all the complicated bits handled for you, then downloading as a simple CSV? Or even simply bulk generating content, such as generating Press Releases from your documents?

All easy with FlowChai :)

I think there's room for many different options in this space, whether that be Personal, Small Business or Enterprise.

Here's an example of automatically scraped arXiv papers on GPT4, turned into a report (with sources) generated by GPT4: https://flowch.ai/shared/6107d220-4e19-4bdc-a566-e84e8a60565...

Learning to leverage LLMs as part of software systems (distinct from LLMs for writing code, that'll be common soon as people catch on and it becomes acceptable to most).

As part of my testing of https://flowch.ai (which we are developing), I've been uploading multiple PDFs (including scientific papers). I then ask a question and ask it to cite its sources as part of the prompt and it'll give the PDFs it's using for the data / quotes it's using. Works well!

It's currently free to use :)

Its integration with public and private data along with layers of automation allowing its use way beyond chat bots. From individual tool to something that can power large parts of a business.

https://flowch.ai - Currently free to use.

The open source solutions have mental overhead and potentially long term maintenance hassle. Even for people who could set it up, for many $20 is worth not having to deal with it.

I want to search based on semantic meaning, rather than partial or exact keyword search.

I'm using pgvector within postgres, so I get all of the postgres benefits.

GPT-4 via ChatGPT, mostly for coding and GPT-4 via https://flowch.ai (our tool) for anything that involves my own data (including my own code), or where I want automation, such as running a prompt on many pieces of data.

Overall this has at least 5x my productivity, possibly 10x.

As well as heavy LLM use, we're also using vectors (with pgvector) for searching text content which is displayed in a table. Works very well.

https://flowch.ai (our project) does this and is currently free to use. It doesn't train a custom model but it gets good results.

A simple way to do this is to upload your files (PDFs, Wod docs, virtually any type is supported), then generate reports using prompts based on those uploaded files. You can go from uploading to results in around a minute.

I've had good results uploading a bunch of documents, then running the same prompt on each of the documents with a few clicks using the "Flow Reports" feature.

We're working on lots of stuff on top of this, like scheduled reports (daily summaries / analysis / newsletters) and automated web scraping and data upload.

Here's an example using a BBC News article that I just uploaded to FlowChai (prompt: summarize in 200 words):

https://flowch.ai/shared/3c6d6ead-3ebc-4190-a143-ffeee81945a...

A lot of LLM libraries seem incredibly overhyped. I've found that I'm able to do a lot with the underlying API calls.

If you're comfortable with doing API calls in your own code, I'd suggest trying to prototype your desired functionality without those libraries to start with.

Thanks for the feedback, will iterate and expand on it.

https://flowch.ai is live and is a hit with early users. As users supply their own prompts regressions in LLMs aren't really an issue, we're currently using both GPT3.5 and GPT-4, both have their place.

Rapidly moved from demos to people actively using it for their day to day work and automation :)

We're taking advantage of new features as they become available, such as OpenAI's functions and larger context windows. Things are evolving quickly!

(and no, we're not using LangChain!)

Simple example of a GPT4 generated report using custom data: https://flowch.ai/shared/73523ec6-4d1d-48a4-bb16-4e9cc01adf1...

A summary of the conversation so far, adding the text of this comment thread to the system: https://flowch.ai/shared/95dd82d1-39d4-4750-b9df-ab2b83cf7ed...

I took the context from those two sources (copy and pasting the text) and it was interesting to see the difference:

GPT 3.5 (fails): https://flowch.ai/shared/90274999-8eaf-4046-9115-7f623264197...

GPT 4 (succeeds): https://flowch.ai/shared/73523ec6-4d1d-48a4-bb16-4e9cc01adf1...