llm-awq
GPTQ-for-LLaMa
llm-awq | GPTQ-for-LLaMa | |
---|---|---|
7 | 75 | |
1,902 | 2,927 | |
10.9% | - | |
8.0 | 8.6 | |
8 days ago | 10 months ago | |
Python | Python | |
MIT License | Apache License 2.0 |
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.
llm-awq
-
TinyChat: Large Language Model on the Edge
TinyChat is an efficient, lightweight, Python-native serving framework for 4-bit LLMs by AWQ. It delivers 2.3x generation speed up on RTX4090.
Code: https://github.com/mit-han-lab/llm-awq/tree/main/tinychat
- FLaNK Stack Weekly 23 Oct 2023
-
New base model InternLM 7B weights released, with 8k context window.
I am having trouble finding any 8bit GPTQ models at all, there don't seem to be any on HF it's almost all 4bit with the odd 3bit of the big ones. Suspect I will have to make my own for eval purposes but it's lower priority on my list then finding a 4bit that's GPU friendly but doesn't have such a performance penalty... Looking at AWQ they have 3 and 4bit versions.
-
Llama33B vs Falcon40B vs MPT30B
Using the currently popular gptq the 3bit quantization hurts performance much more than 4bit, but there's also awq (https://github.com/mit-han-lab/llm-awq) and squishllm (https://github.com/SqueezeAILab/SqueezeLLM) which are able to manage 3bit without as much performance drop - I hope to see them used more commonly.
- New hardware-friendly quantization method
-
Activation-Aware Weight Quantization for LLM Compression Outperforms GPTQ
Better quantization would have a direct and meaningful impact for everyone running local LLMs. The technique has already been applied to both Vicuna and the multimodal LLaMA variant LLaVA.
https://github.com/mit-han-lab/llm-awq
-
New quantization method AWQ outperforms GPTQ in 4-bit and 3-bit with 1.45x speedup and works with multimodal LLMs
GitHub: https://github.com/mit-han-lab/llm-awq
GPTQ-for-LLaMa
-
[P] Early in 2023 I put in a lot of work on a new machine learning project. Now I'm not sure what to do with it.
First I want to make it clear this is not a self promotion post. I hope many machine learning people come at me with questions or comments about this project. A little background about myself. I did work on the 4 bits quantization of LLaMA using GPTQ. (https://github.com/qwopqwop200/GPTQ-for-LLaMa). I've been studying AI in-depth for many years now.
-
GPT-4 Details Leaked
Deploying the 60B version is a challenge though and you might need to apply 4-bit quantization with something like https://github.com/PanQiWei/AutoGPTQ or https://github.com/qwopqwop200/GPTQ-for-LLaMa . Then you can improve the inference speed by using https://github.com/turboderp/exllama .
If you prefer to use an "instruct" model à la ChatGPT (i.e. that does not need few-shot learning to output good results) you can use something like this: https://huggingface.co/TheBloke/Wizard-Vicuna-30B-Uncensored...
-
Rambling
I use gptq-for-llama - from this https://github.com/qwopqwop200/GPTQ-for-LLaMa and Pygmalion 7B.
-
Now that ExLlama is out with reduced VRAM usage, are there any GPTQ models bigger than 7b which can fit onto an 8GB card?
exllama is an optimized implementation of GPTQ-for-LLaMa, allowing you to run 4-bit quantized language models with GPU at great speeds.
-
GGML – AI at the Edge
With a single NVIDIA 3090 and the fastest inference branch of GPTQ-for-LLAMA https://github.com/qwopqwop200/GPTQ-for-LLaMa/tree/fastest-i..., I get a healthy 10-15 tokens per second on the 30B models. IMO GGML is great (And I totally use it) but it's still not as fast as running the models on GPU for now.
-
New quantization method AWQ outperforms GPTQ in 4-bit and 3-bit with 1.45x speedup and works with multimodal LLMs
And exactly what Triton version are they comparing against? I just tried the latest version of this, and on my 4090/12900K I get 77 tokens per second for Llama 7B-128g. My own GPTQ CUDA implementation gets 151 tokens/second on the same model, same hardware. That makes it 96% faster, whereas AWQ is only 79% faster. For 30B-128g I'm currently only getting a 110% speedup over Triton compared to their 178%, but it still seems a little disingenuous to compare against their own CUDA implementation only, when they're trying to present the quantization method as being faster for inference.
-
Introducing Basaran: self-hosted open-source alternative to the OpenAI text completion API
Thanks for the explanation. I think some repos, like text generation webui used gptq for llama (I don't know if it's this repo or another one), anyway most repo that I saw use external things (like gptq for llama)
-
How to use AMD GPU?
cd ../.. git clone https://github.com/qwopqwop200/GPTQ-for-LLaMa.git -b triton cd GPTQ-for-LLaMa pip install -r requirements.txt mkdir -p ../text-generation-webui/repositories ln -s ../../GPTQ-for-LLaMa ../text-generation-webui/repositories/GPTQ-for-LLaMa
-
Help needed with installing quant_cuda for the WebUI
cd repositories git clone https://github.com/qwopqwop200/GPTQ-for-LLaMa pip install -r requirements.txt
-
The installed version of bitsandbytes was compiled without GPU support
# To use the GPTQ models I need to Install GPTQ-for-LLaMa and the monkey patch mkdir repositories cd repositories git clone https://github.com/qwopqwop200/GPTQ-for-LLaMa.git -b triton cd GPTQ-for-LLaMa pip install ninja pip install -r requirements.txt cd cd text-generation-webui # download random model python download-model.py xxx/yyy # try to start the gui python server.py # It returns this warning but it runs bin /home/gm/miniconda3/envs/chat/lib/python3.10/site-packages/bitsandbytes/libbitsandbytes_cpu.so /home/gm/miniconda3/envs/chat/lib/python3.10/site-packages/bitsandbytes/cextension.py:34: UserWarning: The installed version of bitsandbytes was compiled without GPU support. 8-bit optimizers, 8-bit multiplication, and GPU quantization are unavailable. warn("The installed version of bitsandbytes was compiled without GPU support. " /home/gm/miniconda3/envs/chat/lib/python3.10/site-packages/bitsandbytes/libbitsandbytes_cpu.so: undefined symbol: cadam32bit_grad_fp32
What are some alternatives?
SqueezeLLM - [ICML 2024] SqueezeLLM: Dense-and-Sparse Quantization
llama.cpp - LLM inference in C/C++
Voyager - An Open-Ended Embodied Agent with Large Language Models
bitsandbytes - Accessible large language models via k-bit quantization for PyTorch.
langchain4j-examples
text-generation-webui - A Gradio web UI for Large Language Models. Supports transformers, GPTQ, AWQ, EXL2, llama.cpp (GGUF), Llama models.
CML_AMP_AI_Text_Summarization_with_Amazon_Bedrock - CML_AMP_AI_Text_Summarization_with_Amazon_Bedrock
qlora - QLoRA: Efficient Finetuning of Quantized LLMs
kafka-streams-dashboards - showcases Grafana dashboards for Kafka Stream applications leveraging client JMX metrics.
private-gpt - Interact with your documents using the power of GPT, 100% privately, no data leaks
data-in-motion - This is repository for tutorials of Data In Motion starting with Data Distribution
stable-diffusion-webui-docker - Easy Docker setup for Stable Diffusion with user-friendly UI