exllamav2
refact
exllamav2 | refact | |
---|---|---|
17 | 34 | |
3,010 | 1,436 | |
- | 4.2% | |
9.8 | 9.8 | |
3 days ago | 2 days ago | |
Python | JavaScript | |
MIT License | BSD 3-clause "New" or "Revised" License |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
exllamav2
- Running Llama3 Locally
-
Mixture-of-Depths: Dynamically allocating compute in transformers
There are already some implementations out there which attempt to accomplish this!
Here's an example: https://github.com/silphendio/sliced_llama
A gist pertaining to said example: https://gist.github.com/silphendio/535cd9c1821aa1290aa10d587...
Here's a discussion about integrating this capability with ExLlama: https://github.com/turboderp/exllamav2/pull/275
And same as above but for llama.cpp: https://github.com/ggerganov/llama.cpp/issues/4718#issuecomm...
-
What do you use to run your models?
Sorry, I'm somewhat familiar with this term (I've seen it as a model loader in Oobabooga), but still not following the correlation here. Are you saying I should instead be using this project in lieu of llama.cpp? Or are you saying that there is, perhaps, an exllamav2 "extension" or similar within llama.cpp that I can use?
-
I just started having problems with the colab again. I get errors and it just stops. Help?
EDIT: I reported the bug to the exllamav2 Github. It's actually already fixed, just not on any current built release.
-
Yi-34B-200K works on a single 3090 with 47K context/4bpw
install exllamav2 from git with pip install git+https://github.com/turboderp/exllamav2.git. Make sure you have flash attention 2 as well.
-
Tested: ExllamaV2's max context on 24gb with 70B low-bpw & speculative sampling performance
Recent releases for exllamav2 brings working fp8 cache support, which I've been very excited to test. This feature doubles the maximum context length you can run with your model, without any visible downsides.
-
Show HN: Phind Model beats GPT-4 at coding, with GPT-3.5 speed and 16k context
Without batching, I was actually thinking that's kind of modest.
ExllamaV2 will get 48 tokens/s on a 4090, which is much slower/cheaper than an H100:
https://github.com/turboderp/exllamav2#performance
I didn't test codellama, but the 3090 TI figures are in the ballpark of my generation speed on a 3090.
-
Guide for Llama2 70b model merging and exllama2 quantization
First, you need the convert.py script from turboderp's Exllama2 repo. You can read all about the convert.py arguments here.
-
LLM Falcon 180B Needs 720GB RAM to Run
> brute aggressive quantization
Cutting edge quantization like ExLlama's EX2 is far from brute force: https://github.com/turboderp/exllamav2#exl2-quantization
> The format allows for mixing quantization levels within a model to achieve any average bitrate between 2 and 8 bits per weight. Moreover, it's possible to apply multiple quantization levels to each linear layer, producing something akin to sparse quantization wherein more important weights (columns) are quantized with more bits. The same remapping trick that lets ExLlama work efficiently with act-order models allows this mixing of formats to happen with little to no impact on performance. Parameter selection is done automatically by quantizing each matrix multiple times, measuring the quantization error (with respect to the chosen calibration data) for each of a number of possible settings, per layer. Finally, a combination is chosen that minimizes the maximum quantization error over the entire model while meeting a target average bitrate.
Llama.cpp is also working on a feature that let's a small model "guess" the output of a big model which then "checks" it for correctness. This is more of a performance feature, but you could also arrange it to accelerate a big model on a small GPU.
- 70B Llama 2 at 35tokens/second on 4090
refact
- RefactAI: Use best-in-class LLMs for coding in your IDE
-
Supercharge Your Dev Workflow: How Refact's AI-powered Code Completion Boosts Developer Productivity
With over 1.3k stars on GitHub, more than 40k downloads and installs on both VS Code and JetBrains IDEs, and more than 50 positive reviews, it is worth saying that Refact is part of the best product in the AI coding assistant market.
-
What do you use to run your models?
On vscode i sometimes use continue.dev and refact.ai just for fun and they are great!
-
AI Code assistant for about 50-70 users
Refact was made for this: https://github.com/smallcloudai/refact
- Free WebUI for Fine-Tuning and Self-Hosting Open-Source LLMs for Coding
-
LocalPilot: Open-source GitHub Copilot on your MacBook
You should check-out [refact.ai](https://github.com/smallcloudai/refact). It has both autocomplete and chat. It's in active development, with lots of new features coming soon (context search, fine-tuning for larger models, etc)
-
Replit's new AI Model now available on Hugging Face
I don’t recommend that, since that uses the cloud for the actual inference by default (and they provide no guidance for changing that).
I don’t consider cloud inference to count as getting it working “locally” as requested by the comment above yours.
Refact works nicely and works locally, but the challenge with any new model is making it be supported by the existing software: https://github.com/smallcloudai/refact/
- Refact.ai 1.0.0 Released
-
📝 🚀 Creating our first documentation from scratch using Astro and Refact AI coding assistant
Previously, we used Astro for our refact.ai website and wanted to stay within the Astro ecosystem for the documentation.
-
🤖We trained a small 1.6b code model and you can use it as a personal copilot in Refact for free🤖
Refact LLM can be easily integrated into existing developers workflows with an open-source docker container and VS Code and JetBrains plugins. With Refact's intuitive user interface, developers can utilize the model easily for a variety of coding tasks. Finetune is available in the self-hosting (docker) and Enterprise versions, making suggestions more relevant for your private codebase.
What are some alternatives?
llama.cpp - LLM inference in C/C++
tabby - Self-hosted AI coding assistant
koboldcpp - A simple one-file way to run various GGML and GGUF models with KoboldAI's UI
fauxpilot - FauxPilot - an open-source alternative to GitHub Copilot server
SillyTavern - LLM Frontend for Power Users.
ollama - Get up and running with Llama 3, Mistral, Gemma, and other large language models.
ChatGPT-AutoExpert - 🚀🧠💬 Supercharged Custom Instructions for ChatGPT (non-coding) and ChatGPT Advanced Data Analysis (coding).
llama-cpp-python - Python bindings for llama.cpp
OmniQuant - [ICLR2024 spotlight] OmniQuant is a simple and powerful quantization technique for LLMs.
developer - the first library to let you embed a developer agent in your own app!
BlockMerge_Gradient - Merge Transformers language models by use of gradient parameters.
supervision - We write your reusable computer vision tools. 💜