|
--- |
|
license: gpl |
|
inference: false |
|
--- |
|
|
|
# gpt4-x-vicuna-13B-GPTQ |
|
|
|
This repo contains 4bit GPTQ format quantised models of [NousResearch's gpt4-x-vicuna-13b](https://huggingface.co/NousResearch/gpt4-x-vicuna-13b). |
|
|
|
It is the result of quantising to 4bit using [GPTQ-for-LLaMa](https://github.com/qwopqwop200/GPTQ-for-LLaMa). |
|
|
|
## Repositories available |
|
|
|
* [4bit GPTQ models for GPU inference](https://huggingface.co/TheBloke/gpt4-x-vicuna-13B-GPTQ). |
|
* [4bit and 5bit GGML models for CPU inference](https://huggingface.co/TheBloke/gpt4-x-vicuna-13B-GGML). |
|
* [float16 models in HF format for GPU inference](https://huggingface.co/TheBloke/gpt4-x-vicuna-13B-HF). |
|
|
|
## Provided files |
|
| Name | Quant method | Bits | Size | RAM required | Use case | |
|
| ---- | ---- | ---- | ---- | ---- | ----- | |
|
`gpt4-x-vicuna-13B.ggml.q4_0.bin` | q4_0 | 4bit | 8.14GB | 10GB | Maximum compatibility | |
|
`gpt4-x-vicuna-13B.ggml.q4_2.bin` | q4_2 | 4bit | 8.14GB | 10GB | Best compromise between resources, speed and quality | |
|
`gpt4-x-vicuna-13B.ggml.q5_0.bin` | q5_0 | 5bit | 8.95GB | 11GB | Brand new 5bit method. Potentially higher quality than 4bit, at cost of slightly higher resources. | |
|
`gpt4-x-vicuna-13B.ggml.q5_1.bin` | q5_1 | 5bit | 9.76GB | 12GB | Brand new 5bit method. Slightly higher resource usage than q5_0.| |
|
|
|
* The q4_0 file provides lower quality, but maximal compatibility. It will work with past and future versions of llama.cpp |
|
* The q4_2 file offers the best combination of performance and quality. This format is still subject to change and there may be compatibility issues, see below. |
|
* The q5_0 file is using brand new 5bit method released 26th April. This is the 5bit equivalent of q4_0. |
|
* The q5_1 file is using brand new 5bit method released 26th April. This is the 5bit equivalent of q4_1. |
|
|
|
## q4_2 compatibility |
|
|
|
q4_2 is a relatively new 4bit quantisation method offering improved quality. However they are still under development and their formats are subject to change. |
|
|
|
In order to use these files you will need to use recent llama.cpp code. And it's possible that future updates to llama.cpp could require that these files are re-generated. |
|
|
|
If and when the q4_2 file no longer works with recent versions of llama.cpp I will endeavour to update it. |
|
|
|
If you want to ensure guaranteed compatibility with a wide range of llama.cpp versions, use the q4_0 file. |
|
|
|
## q5_0 and q5_1 compatibility |
|
|
|
These new methods were released to llama.cpp on 26th April. You will need to pull the latest llama.cpp code and rebuild to be able to use them. |
|
|
|
Don't expect any third-party UIs/tools to support them yet. |
|
|
|
## How to run in `llama.cpp` |
|
|
|
I use the following command line; adjust for your tastes and needs: |
|
|
|
``` |
|
./main -t 12 -m gpt4-x-vicuna-13B.ggml.q4_2.bin --color -c 2048 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "Below is an instruction that describes a task. Write a response that appropriately completes the request. |
|
### Instruction: |
|
Write a story about llamas |
|
### Response:" |
|
``` |
|
Change `-t 12` to the number of physical CPU cores you have. For example if your system has 8 cores/16 threads, use `-t 8`. |
|
|
|
If you want to have a chat-style conversation, replace the `-p <PROMPT>` argument with `-i -ins` |
|
|
|
## How to run in `text-generation-webui` |
|
|
|
Further instructions here: [text-generation-webui/docs/llama.cpp-models.md](https://github.com/oobabooga/text-generation-webui/blob/main/docs/llama.cpp-models.md). |
|
|
|
Note: at this time text-generation-webui will not support the new q5 quantisation methods. |
|
|
|
**Thireus** has written a [great guide on how to update it to the latest llama.cpp code](https://huggingface.co/TheBloke/wizardLM-7B-GGML/discussions/5) so that these files can be used in the UI. |
|
|
|
# Original model card |
|
|
|
As a base model used https://huggingface.co/eachadea/vicuna-13b-1.1 |
|
|
|
Finetuned on Teknium's GPTeacher dataset, unreleased Roleplay v2 dataset, GPT-4-LLM dataset, and Nous Research Instruct Dataset |
|
|
|
Approx 180k instructions, all from GPT-4, all cleaned of any OpenAI censorship/"As an AI Language Model" etc. |
|
|
|
Base model still has OpenAI censorship. Soon, a new version will be released with cleaned vicuna from https://huggingface.co/datasets/anon8231489123/ShareGPT_Vicuna_unfiltere |
|
|
|
Trained on 8 A100-80GB GPUs for 5 epochs following Alpaca deepspeed training code. |
|
|
|
Nous Research Instruct Dataset will be released soon. |
|
|
|
GPTeacher, Roleplay v2 by https://huggingface.co/teknium |
|
|
|
Wizard LM by https://github.com/nlpxucan |
|
|
|
Nous Research Instruct Dataset by https://huggingface.co/karan4d and https://huggingface.co/huemin |
|
|
|
Compute provided by our project sponsor https://redmond.ai/ |
