FastChat
bitsandbytes
Our great sponsors
FastChat | bitsandbytes | |
---|---|---|
82 | 61 | |
33,464 | 5,299 | |
4.8% | - | |
9.7 | 9.4 | |
6 days ago | 7 days ago | |
Python | Python | |
Apache License 2.0 | MIT 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.
FastChat
-
LLMs on your local Computer (Part 1)
FastChat
- FLaNK AI for 11 March 2024
- FLaNK 04 March 2024
- ChatGPT for Teams
-
LM Studio – Discover, download, and run local LLMs
How does it compare with something like FastChat? https://github.com/lm-sys/FastChat
Feature set seems like a decent amount of overlap. One limitation of FastChat, as far as I can tell, is that one is limited to the models that FastChat supports (though I think it would be minor to modify it to support arbitrary models?)
-
Video-LLaVA
Looks like the Vicuna repo is Apache 2.0 also[1].
What's the interpretation of copyright law that would prevent the code being Apache 2.0 based on the source of the fine-tuning dataset?
-
🔥🚀 Top 10 Open-Source Must-Have Tools for Crafting Your Own Chatbot 🤖💬
Check how to start with FastChat. Support FastChat on GitHub ⭐
-
Show HN: ChatAPI – PWA to Use ChatGPT by API Build with Alpine.js
For something a little heavier but much more robust in terms of features/functionality I've been enjoying FastChat: https://github.com/lm-sys/FastChat
It allows you to plug in different backends so that you can use OpenAI compatible clients with various LLM's, selfhosted or otherwise.
- FLaNK Stack Weekly 09 Oct 2023
-
Show HN: Open-source proxy server for Llama2, GPT-4, Claude2 with Logging,Cache
If you do want to self-host - there's some great libraries like https://github.com/lm-sys/FastChat and https://github.com/ggerganov/llama.cpp that might be helpful
If none of these really solve your issue - feel free to email me and I'm happy to help you figure something out - [email protected]
bitsandbytes
-
French AI startup Mistral secures €2B valuation
No. Without the inference code, the best we can have are guesses on its implementation, so the benchmark figures we can get could be quite wrong. It does seem better than Llama2-70B in my tests, which rely on the work done by Dmytro Dzhulgakov[0] and DiscoResearch[1].
But the point of releasing on bittorrent is to see the effervescence in hobbyist research and early attempts at MoE quantization, which are already ongoing[2]. They are benefitting from the community.
[0]: https://github.com/dzhulgakov/llama-mistral
[1]: https://huggingface.co/DiscoResearch/mixtral-7b-8expert
[2]: https://github.com/TimDettmers/bitsandbytes/tree/sparse_moe
-
Lora training with Kohya issue
CUDA SETUP: To manually override the PyTorch CUDA version please see:https://github.com/TimDettmers/bitsandbytes/blob/main/how_to_use_nonpytorch_cuda.md
- FLaNK Stack Weekly for 30 Oct 2023
-
A comprehensive guide to running Llama 2 locally
While on the surface, a 192GB Mac Studio seems like a great deal (it's not much more than a 48GB A6000!), there are several reasons why this might not be a good idea:
* I assume most people have never used llama.cpp Metal w/ large models. It will drop to CPU speeds whenever the context window is full: https://github.com/ggerganov/llama.cpp/issues/1730#issuecomm... - while sure this might be fixed in the future, it's been an issue since Metal support was added, and is a significant problem if you are actually trying to actually use it for inferencing. With 192GB of memory, you could probably run larger models w/o quantization, but I've never seen anyone post benchmarks of their experiences. Note that at that point, the limited memory bandwidth will be a big factor.
* If you are planning on using Apple Silicon for ML/training, I'd also be wary. There are multi-year long open bugs in PyTorch[1], and most major LLM libs like deepspeed, bitsandbytes, etc don't have Apple Silicon support[2][3].
You can see similar patterns w/ Stable Diffusion support [4][5] - support lagging by months, lots of problems and poor performance with inference, much less fine tuning. You can apply this to basically any ML application you want (srt, tts, video, etc)
Macs are fine to poke around with, but if you actually plan to do more than run a small LLM and say "neat", especially for a business, recommending a Mac for anyone getting started w/ ML workloads is a bad take. (In general, for anyone getting started, unless you're just burning budget, renting cloud GPU is going to be the best cost/perf, although on-prem/local obviously has other advantages.)
[1] https://github.com/pytorch/pytorch/issues?q=is%3Aissue+is%3A...
[2] https://github.com/microsoft/DeepSpeed/issues/1580
[3] https://github.com/TimDettmers/bitsandbytes/issues/485
[4] https://github.com/AUTOMATIC1111/stable-diffusion-webui/disc...
[5] https://forums.macrumors.com/threads/ai-generated-art-stable...
-
Considering getting a Jetson AGX Orin.. anyone have experience with it?
Do you by chance have any details on how to run oobagooba on the Orin? I keep running into this issue seemingly related to bitsandbytes.
-
Finetuning on multiple GPUs
If it also has QLoRA that would be the best but afaik it's not implemented in bitsandbytes yet?
- A new paper has been released, QLoRA, which is nothing short of game-changing for the ability to train and fine-tune LLMs on consumers' GPUs.
-
Anybody tried Lion: Adversarial Distillation of Closed-Source Large Language Model?
After looking in the bitsandbytes github i wanted to understand what the Added PagedLion and bf16 Lion. means :)
-
QLoRA: Efficient Finetuning of Quantized LLMs
Tim Dettmers is such a star. He's probably done more to make low-resource LLMs usable than anyone else.
First bitsandbytes[1] and now this.
What are some alternatives?
GPTQ-for-LLaMa - 4 bits quantization of LLaMA using GPTQ
text-generation-webui - A Gradio web UI for Large Language Models. Supports transformers, GPTQ, AWQ, EXL2, llama.cpp (GGUF), Llama models.
llama.cpp - LLM inference in C/C++
accelerate - 🚀 A simple way to launch, train, and use PyTorch models on almost any device and distributed configuration, automatic mixed precision (including fp8), and easy-to-configure FSDP and DeepSpeed support
gpt4all - gpt4all: run open-source LLMs anywhere
LocalAI - :robot: The free, Open Source OpenAI alternative. Self-hosted, community-driven and local-first. Drop-in replacement for OpenAI running on consumer-grade hardware. No GPU required. Runs gguf, transformers, diffusers and many more models architectures. It allows to generate Text, Audio, Video, Images. Also with voice cloning capabilities.
llama-cpp-python - Python bindings for llama.cpp
Dreambooth-Stable-Diffusion-cpu - Implementation of Dreambooth (https://arxiv.org/abs/2208.12242) with Stable Diffusion
alpaca.cpp - Locally run an Instruction-Tuned Chat-Style LLM
mlc-llm - Enable everyone to develop, optimize and deploy AI models natively on everyone's devices.
qlora - QLoRA: Efficient Finetuning of Quantized LLMs