Share this page!

Enter Twitter Thread URL to Unroll

Needs to be the full URL like: https://twitter.com/threadreaderapp/status/1644127596119195649

How to get URL link on Twitter App

  1. On the Twitter thread, click on or icon on the bottom
  2. Click again on or Share Via icon
  3. Click on Copy Link to Tweet
  4. Paste it above and click "Unroll Thread"!
  5. More info at Twitter Help
GPT Index (LlamaIndex 🦙) Profile picture
@gpt_index
Apr 24 • 3 tweets • 2 min read Twitter logo Read on Twitter
You can now `pip install llama-agi`! 🛠️🧠

Get access to a core set of abstractions/tools that allow you to easily build #babyAGI/#AutoGPT-like capabilities yourself 🔥

Example notebook here, it’s super easy to get started: colab.research.google.com/drive/1yXengyQ…
The core abstractions:
✅ Task managers - add/generate/parse/prioritize tasks using @gpt_index modules
🤖 Execution Agents - execute actions with the help of @LangChainAI agent abstractions
🏃AGI Runner - call agents/task managers in an outer loop!

Docs coming soon!
The goal of llama-agi (part of our Llama Lab repo), is to provide a fun playground for building experimental AGI capabilities.

Can also help to inform core features we put in llama-index.

Link: github.com/run-llama/llam…

We’re very open to contributions and new ideas too 💡

• • •

Missing some Tweet in this thread? You can try to force a refresh
　

Keep Current with GPT Index (LlamaIndex 🦙)

GPT Index (LlamaIndex 🦙) Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @gpt_index

GPT Index (LlamaIndex 🦙) Profile picture
Apr 26
LlamaIndex 0.5.25 brings some awesome new integrations:

🗳️ MyScale data loader + Vector Store 👇
🏞️ BLIP/BLIP2 image captioning data loaders 👇
📘Jupyter ipynb data loaders
🤗 @huggingface FS data loader
⚙️Evaporate structured data extractor 👇

See details below!
MyScale is an AI database that can manage both structured and vectorized data.

Main page: myscale.com

It’s implemented as both a data loader + vector index within @gpt_index.

gpt-index.readthedocs.io/en/latest/how_…
BLIP/BLIP2 provide image captioning capabilities. Now you have even more powerful tools to convert image -> text and ingest within @gpt_index!

llamahub.ai/l/file-image_b…
llamahub.ai/l/file-image_b…
Read 5 tweets
GPT Index (LlamaIndex 🦙) Profile picture
Mar 10
LlamaIndex 0.4.25 🦙: some big features from the community 🥳
- ⚙️ An *optimizer* to help reduce your LLM token usage by up to 50% or more! (s/o @shi_hongyi) more details below! 👇
- ♻️ You can now *sync* Document updates with an index! (s/o Logan), see below 👇
(1) To optimize LLM token usage: specify a `SentenceEmbeddingOptimizer` when querying a vector index.

We “strip-out” sentences from a text chunk that have low embedding similarity, before the LLM call.

See 🖼️ below - we cut LLM token usage by 50% ImageImage
(2) You can now sync updates from your source Documents to an index - we’ll only update the index for Documents that have changed.

Just call `index.refresh` on the new Documents! Image
Read 4 tweets
GPT Index (LlamaIndex 🦙) Profile picture
Feb 28
LlamaIndex 0.4.15:
- Biggest update is that we’ve fixed the composability over vector stores! See below for a demonstration of the new UX flow. 👇
- Added @weaviate_io multi-threaded batch importing to decrease load time (thanks @MohdShukriHasa1 for suggestion!)
A big pain point with the composability UX was that it didn’t work fully work over external vector stores.

Why? Because configs were keyed by type, not on ID.

So if you have two Pinecone/Weaviate stores, you couldn’t distinguish them during query-time!

Now that’s fixed.
The biggest change here is to set an ID in the query config when defining a graph; this ties the config to a specific index!

You can set an id for each index through `set_doc_id`. ImageImage
Read 4 tweets
GPT Index (LlamaIndex 🦙) Profile picture
Feb 17
Today, we’re kicking off a rebrand of @gpt_index to : 🎉🦙 LlamaIndex 🦙🎉

🦙 LlamaIndex + 🦙 LlamaHub aim to provide the central interface between LLM’s and your data.
This will be a *gradual* process. We’re starting this off by uploading a `llama-index` pip package you can now use (don’t worry, `pip install gpt-index` is still there!)

llama-index is an exact duplicate of gpt-index (just `import llama_index` instead of `import gpt_index`)
README’s are not updated yet (sorry about that!), but package is here: pypi.org/project/llama-….

Over the course of the next few weeks, we’ll start to update the docs/readmes and classes (while still supporting backwards compat).
Read 4 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us on Twitter!

Send Email!

:(