GGUF is a new format introduced by the llama.cpp team on August 21st 2023. It is a replacement for GGML, which is no longer supported by llama.cpp. GGUF offers numerous advantages over GGML, such as better tokenisation, and support for special tokens. It is also supports metadata, and is designed to be extensible.

Here is an incomplate list of clients and libraries that are known to support GGUF:

llama.cpp. The source project for GGUF. Offers a CLI and a server option.
text-generation-webui, the most widely used web UI, with many features and powerful extensions. Supports GPU acceleration.
KoboldCpp, a fully featured web UI, with GPU accel across all platforms and GPU architectures. Especially good for story telling.
LM Studio, an easy-to-use and powerful local GUI for Windows and macOS (Silicon), with GPU acceleration.
LoLLMS Web UI, a great web UI with many interesting and unique features, including a full model library for easy model selection.
Faraday.dev, an attractive and easy to use character-based chat GUI for Windows and macOS (both Silicon and Intel), with GPU acceleration.
ctransformers, a Python library with GPU accel, LangChain support, and OpenAI-compatible AI server.
llama-cpp-python, a Python library with GPU accel, LangChain support, and OpenAI-compatible API server.
candle, a Rust ML framework with a focus on performance, including GPU support, and ease of use.

Repositories available

Prompt template: CodeFuse

<|role_start|>system<|role_end|>{system_message}
<|role_start|>human<|role_end|>{prompt}
<|role_start|>bot<|role_end|>

Licensing

The creator of the source model has listed its license as other, and this quantization has therefore used that same license.

As this model is based on Llama 2, it is also subject to the Meta Llama 2 license terms, and the license files for that are additionally included. It should therefore be considered as being claimed to be licensed under both licenses. I contacted Hugging Face for clarification on dual licensing but they do not yet have an official position. Should this change, or should Meta provide any feedback on this situation, I will update this section accordingly.

In the meantime, any questions regarding licensing, and in particular how these two licenses might interact, should be directed to the original model repository: CodeFuse AI's CodeFuse CodeLlama 34B.

Compatibility

These quantised GGUFv2 files are compatible with llama.cpp from August 27th onwards, as of commit d0cee0d36d5be95a0d9088b674dbb27354107221

They are also compatible with many third party UIs and libraries - please see the list at the top of this README.

Explanation of quantisation methods

Click to see details

The new methods available are:

GGML_TYPE_Q2_K - "type-1" 2-bit quantization in super-blocks containing 16 blocks, each block having 16 weight. Block scales and mins are quantized with 4 bits. This ends up effectively using 2.5625 bits per weight (bpw)
GGML_TYPE_Q3_K - "type-0" 3-bit quantization in super-blocks containing 16 blocks, each block having 16 weights. Scales are quantized with 6 bits. This end up using 3.4375 bpw.
GGML_TYPE_Q4_K - "type-1" 4-bit quantization in super-blocks containing 8 blocks, each block having 32 weights. Scales and mins are quantized with 6 bits. This ends up using 4.5 bpw.
GGML_TYPE_Q5_K - "type-1" 5-bit quantization. Same super-block structure as GGML_TYPE_Q4_K resulting in 5.5 bpw
GGML_TYPE_Q6_K - "type-0" 6-bit quantization. Super-blocks with 16 blocks, each block having 16 weights. Scales are quantized with 8 bits. This ends up using 6.5625 bpw

Refer to the Provided Files table below to see what files use which methods, and how.

Provided files

Name	Quant method	Bits	Size	Max RAM required	Use case
codefuse-codellama-34b.Q2_K.gguf	Q2_K	2	14.21 GB	16.71 GB	smallest, significant quality loss - not recommended for most purposes
codefuse-codellama-34b.Q3_K_S.gguf	Q3_K_S	3	14.61 GB	17.11 GB	very small, high quality loss
codefuse-codellama-34b.Q3_K_M.gguf	Q3_K_M	3	16.28 GB	18.78 GB	very small, high quality loss
codefuse-codellama-34b.Q3_K_L.gguf	Q3_K_L	3	17.77 GB	20.27 GB	small, substantial quality loss
codefuse-codellama-34b.Q4_0.gguf	Q4_0	4	19.05 GB	21.55 GB	legacy; small, very high quality loss - prefer using Q3_K_M
codefuse-codellama-34b.Q4_K_S.gguf	Q4_K_S	4	19.15 GB	21.65 GB	small, greater quality loss
codefuse-codellama-34b.Q4_K_M.gguf	Q4_K_M	4	20.22 GB	22.72 GB	medium, balanced quality - recommended
codefuse-codellama-34b.Q5_0.gguf	Q5_0	5	23.24 GB	25.74 GB	legacy; medium, balanced quality - prefer using Q4_K_M
codefuse-codellama-34b.Q5_K_S.gguf	Q5_K_S	5	23.24 GB	25.74 GB	large, low quality loss - recommended
codefuse-codellama-34b.Q5_K_M.gguf	Q5_K_M	5	23.84 GB	26.34 GB	large, very low quality loss - recommended
codefuse-codellama-34b.Q6_K.gguf	Q6_K	6	27.68 GB	30.18 GB	very large, extremely low quality loss
codefuse-codellama-34b.Q8_0.gguf	Q8_0	8	35.86 GB	38.36 GB	very large, extremely low quality loss - not recommended

Note: the above RAM figures assume no GPU offloading. If layers are offloaded to the GPU, this will reduce RAM usage and use VRAM instead.

How to download GGUF files

Note for manual downloaders: You almost never want to clone the entire repo! Multiple different quantisation formats are provided, and most users only want to pick and download a single file.

The following clients/libraries will automatically download models for you, providing a list of available models to choose from:

LM Studio
LoLLMS Web UI
Faraday.dev

In `text-generation-webui`

Under Download Model, you can enter the model repo: TheBloke/CodeFuse-CodeLlama-34B-GGUF and below it, a specific filename to download, such as: codefuse-codellama-34b.q4_K_M.gguf.

Then click Download.

On the command line, including multiple files at once

I recommend using the huggingface-hub Python library:

pip3 install huggingface-hub>=0.17.1

Then you can download any individual model file to the current directory, at high speed, with a command like this:

huggingface-cli download TheBloke/CodeFuse-CodeLlama-34B-GGUF codefuse-codellama-34b.q4_K_M.gguf --local-dir . --local-dir-use-symlinks False

More advanced huggingface-cli download usage

You can also download multiple files at once with a pattern:

huggingface-cli download TheBloke/CodeFuse-CodeLlama-34B-GGUF --local-dir . --local-dir-use-symlinks False --include='*Q4_K*gguf'

For more documentation on downloading with huggingface-cli, please see: HF -> Hub Python Library -> Download files -> Download from the CLI.

To accelerate downloads on fast connections (1Gbit/s or higher), install hf_transfer:

pip3 install hf_transfer

And set environment variable HF_HUB_ENABLE_HF_TRANSFER to 1:

HUGGINGFACE_HUB_ENABLE_HF_TRANSFER=1 huggingface-cli download TheBloke/CodeFuse-CodeLlama-34B-GGUF codefuse-codellama-34b.q4_K_M.gguf --local-dir . --local-dir-use-symlinks False

Windows CLI users: Use set HUGGINGFACE_HUB_ENABLE_HF_TRANSFER=1 before running the download command.

Example `llama.cpp` command

Make sure you are using llama.cpp from commit d0cee0d36d5be95a0d9088b674dbb27354107221 or later.

./main -ngl 32 -m codefuse-codellama-34b.q4_K_M.gguf --color -c 4096 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "<|role_start|>system<|role_end|>{system_message}\n<|role_start|>human<|role_end|>{prompt}\n<|role_start|>bot<|role_end|>"

Change -ngl 32 to the number of layers to offload to GPU. Remove it if you don't have GPU acceleration.

Change -c 4096 to the desired sequence length. For extended sequence models - eg 8K, 16K, 32K - the necessary RoPE scaling parameters are read from the GGUF file and set by llama.cpp automatically.

If you want to have a chat-style conversation, replace the -p <PROMPT> argument with -i -ins

For other parameters and how to use them, please refer to the llama.cpp documentation

How to run in `text-generation-webui`

Further instructions here: text-generation-webui/docs/llama.cpp.md.

How to run from Python code

You can use GGUF models from Python using the llama-cpp-python or ctransformers libraries.

How to load this model from Python using ctransformers

First install the package

bash

# Base ctransformers with no GPU acceleration
pip install ctransformers>=0.2.24
# Or with CUDA GPU acceleration
pip install ctransformers[cuda]>=0.2.24
# Or with ROCm GPU acceleration
CT_HIPBLAS=1 pip install ctransformers>=0.2.24 --no-binary ctransformers
# Or with Metal GPU acceleration for macOS systems
CT_METAL=1 pip install ctransformers>=0.2.24 --no-binary ctransformers

Simple example code to load one of these GGUF models

python

from ctransformers import AutoModelForCausalLM

# Set gpu_layers to the number of layers to offload to GPU. Set to 0 if no GPU acceleration is available on your system.
llm = AutoModelForCausalLM.from_pretrained("TheBloke/CodeFuse-CodeLlama-34B-GGUF", model_file="codefuse-codellama-34b.q4_K_M.gguf", model_type="llama", gpu_layers=50)

print(llm("AI is going to"))

How to use with LangChain

Here's guides on using llama-cpp-python or ctransformers with LangChain:

Discord

For further support, and discussions on these models and AI in general, join us at:

TheBloke AI's Discord server

Thanks, and how to contribute

Thanks to the chirper.ai team!

Thanks to Clay from gpus.llm-utils.org!

I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.

If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.

Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.

Patreon: https://patreon.com/TheBlokeAI
Ko-Fi: https://ko-fi.com/TheBlokeAI

Special thanks to: Aemon Algiz.

Patreon special mentions: Alicia Loh, Stephen Murray, K, Ajan Kanaga, RoA, Magnesian, Deo Leter, Olakabola, Eugene Pentland, zynix, Deep Realms, Raymond Fosdick, Elijah Stavena, Iucharbius, Erik Bjäreholt, Luis Javier Navarrete Lozano, Nicholas, theTransient, John Detwiler, alfie_i, knownsqashed, Mano Prime, Willem Michiel, Enrico Ros, LangChain4j, OG, Michael Dempsey, Pierre Kircher, Pedro Madruga, James Bentley, Thomas Belote, Luke @flexchar, Leonard Tan, Johann-Peter Hartmann, Illia Dulskyi, Fen Risland, Chadd, S_X, Jeff Scroggin, Ken Nordquist, Sean Connelly, Artur Olbinski, Swaroop Kallakuri, Jack West, Ai Maven, David Ziegler, Russ Johnson, transmissions 11, John Villwock, Alps Aficionado, Clay Pascal, Viktor Bowallius, Subspace Studios, Rainer Wilmers, Trenton Dambrowitz, vamX, Michael Levine, 준교 김, Brandon Frisco, Kalila, Trailburnt, Randy H, Talal Aujan, Nathan Dryer, Vadim, 阿明, ReadyPlayerEmma, Tiffany J. Kim, George Stoitzev, Spencer Kim, Jerry Meng, Gabriel Tamborski, Cory Kujawski, Jeffrey Morgan, Spiking Neurons AB, Edmond Seymore, Alexandros Triantafyllidis, Lone Striker, Cap'n Zoog, Nikolai Manek, danny, ya boyyy, Derek Yates, usrbinkat, Mandus, TL, Nathan LeClaire, subjectnull, Imad Khwaja, webtim, Raven Klaugh, Asp the Wyvern, Gabriel Puliatti, Caitlyn Gatomon, Joseph William Delisle, Jonathan Leane, Luke Pendergrass, SuperWojo, Sebastain Graf, Will Dee, Fred von Graf, Andrey, Dan Guido, Daniel P. Andersen, Nitin Borwankar, Elle, Vitor Caleffi, biorpg, jjj, NimbleBox.ai, Pieter, Matthew Berman, terasurfer, Michael Davis, Alex, Stanislav Ovsiannikov

Thank you to all my generous patrons and donaters!

And thank you again to a16z for their generous grant.

Original model card: CodeFuse AI's CodeFuse CodeLlama 34B

Model Card for CodeFuse-CodeLlama-34B

logo

[中文] [English]

Model Description

CodeFuse-CodeLlama-34B is a 34B Code-LLM finetuned by QLoRA of multiple code tasks（600k instrunctions/answers） on the base model CodeLlama-34b-Python. The context length of finetuning is 4K while it is able to be finetuned by 16k context if necessary.

News and Updates

🔥🔥🔥 CodeFuse-CodeLlama34B-MFT has achived 74.4% of pass@1 on HumanEval, which is SOTA at present.

Code Community

Homepage: 🏡 https://github.com/codefuse-ai (Please give us your support with a Star🌟 + Fork🚀 + Watch👀)

If you wish to fine-tune the model yourself, you can visit ✨MFTCoder✨✨
If you wish to deploy the model yourself, you can visit ✨FasterTransformer4CodeFuse✨✨
If you wish to see a demo of the model, you can visit ✨CodeFuse Demo✨✨

Performance

Model	HumanEval(pass@1)	Date
CodeFuse-CodeLlama-34B	74.4%	2023.9
WizardCoder-Python-34B-V1.0	73.2%	2023.8
GPT-4(zero-shot)	67.0%	2023.3
PanGu-Coder2 15B	61.6%	2023.8
CodeLlama-34b-Python	53.7%	2023.8
CodeLlama-34b	48.8%	2023.8
GPT-3.5(zero-shot)	48.1%	2022.11
OctoCoder	46.2%	2023.8
StarCoder-15B	33.6%	2023.5
LLaMA 2 70B(zero-shot)	29.9%	2023.7

Requirements

python>=3.8
pytorch>=2.0.0
transformers==4.32.0
Sentencepiece
CUDA 11.4

Inference String Format

The inference string is a concatenated string formed by combining conversation data(system, human and bot contents) in the training data format. It is used as input during the inference process. Here is an example format of the concatenated string:

python

"""
<|role_start|>system<|role_end|>System instruction
<|role_start|>human<|role_end|>Human 1st round input
<|role_start|>bot<|role_end|>Bot 1st round output</s>
<|role_start|>human<|role_end|>Human 2nd round input
<|role_start|>bot<|role_end|>Bot 2nd round output</s>
...
...
...
<|role_start|>human<|role_end|>Human nth round input
<|role_start|>bot<|role_end|>{Bot output to be genreated}</s>
"""

When applying inference, you always make your input string end with "<|role_start|>bot<|role_end|>" to ask the model generating answers.

Quickstart

pip install -r requirements.txt

python

import torch
from transformers import (
    AutoTokenizer,
    AutoModelForCausalLM,
)
tokenizer = AutoTokenizer.from_pretrained(mode_name_or_path, trust_remote_code=True, use_fast=False, legacy=False)
tokenizer.padding_side = "left"
tokenizer.pad_token_id = tokenizer.convert_tokens_to_ids("<unk>")
tokenizer.eos_token_id = tokenizer.convert_tokens_to_ids("</s>")
# try 4bit loading if cuda memory not enough
model = AutoModelForCausalLM.from_pretrained(mode_name_or_path,
                                             trust_remote_code=True,
                                             load_in_4bit=False,
                                             device_map="auto",
                                             torch_dtype=torch.bfloat16)
model.eval()

HUMAN_ROLE_START_TAG = "<|role_start|>human<|role_end|>"
BOT_ROLE_START_TAG = "<|role_start|>bot<|role_end|>"

text = f"{HUMAN_ROLE_START_TAG}write a python function of quick sort.{BOT_ROLE_START_TAG}"
inputs = tokenizer(text, return_tensors='pt', padding=True, add_special_tokens=False).to("cuda")
outputs = model.generate(
        inputs=inputs["input_ids"],
        attention_mask=inputs["attention_mask"],
        max_new_tokens=512,
        top_p=0.95,
        temperature=0.1,
        do_sample=True,
        eos_token_id=tokenizer.eos_token_id,
        pad_token_id=tokenizer.pad_token_id
    )
gen_text = tokenizer.batch_decode(outputs[:, inputs["input_ids"].shape[1]:], skip_special_tokens=True)
print(gen_text)

MD5

We notice that the file may be corrupted during transfer process. Please check MD5 value before use.

Model File	MD5 Value
pytorch_model-00001-of-00007.bin	8d544b1bcb3449934184d4141137329c
pytorch_model-00002-of-00007.bin	9d5dbb30911e48a42fb6d0fcabb322a4
pytorch_model-00003-of-00007.bin	b0d4aecee0457d9332005a187e1fffed
pytorch_model-00004-of-00007.bin	5c7e002de5eab77d0194a2b0f6de0c24
pytorch_model-00005-of-00007.bin	d22a511aa26b5b17117b665a877490ab
pytorch_model-00006-of-00007.bin	a5c28ac277fac07d16dd66537e54d109
pytorch_model-00007-of-00007.bin	a967e2c6195477b7407089c0bffa2d53

模型简介

CodeFuse-CodeLlama34B-MFT 是一个通过QLoRA对基座模型CodeLlama-34b-Python进行多代码任务微调的代码大模型。模型微调采用了4k上下文。如果有必要，可以扩展到16k。

新闻

🔥🔥🔥 CodeFuse-CodeLlama34B-MFT模型在HumanEval pass@1上可以达到74.4%, 为当前开源SOTA。

代码社区

大本营： 🏡 https://github.com/codefuse-ai （欢迎为我们的项目一键三连 Star🌟 + Fork🚀 + Watch👀）

如果您想自己微调该模型，可以访问 ✨MFTCoder✨✨
如果您想自己部署该模型，可以访问 ✨FasterTransformer4CodeFuse✨✨
如果您想观看该模型示例，可以访问 ✨CodeFuse Demo✨✨

评测表现(代码)

模型	HumanEval(pass@1)	日期
CodeFuse-CodeLlama-34B	74.4%	2023.9
WizardCoder-Python-34B-V1.0	73.2%	2023.8
GPT-4(zero-shot)	67.0%	2023.3
PanGu-Coder2 15B	61.6%	2023.8
CodeLlama-34b-Python	53.7%	2023.8
CodeLlama-34b	48.8%	2023.8
GPT-3.5(zero-shot)	48.1%	2022.11
OctoCoder	46.2%	2023.8
StarCoder-15B	33.6%	2023.5
LLaMA 2 70B(zero-shot)	29.9%	2023.7

Requirements

python>=3.8
pytorch>=2.0.0
transformers==4.32.0
CUDA 11.4

推理数据格式

推理数据为模型在训练数据格式下拼接的字符串形式，它也是推理时输入prompt拼接的方式：

python

"""
<|role_start|>system<|role_end|>这是System指令
<|role_start|>human<|role_end|>这是第1轮用户输入的问题
<|role_start|>bot<|role_end|>这是第1轮模型生成的内容</s>
<|role_start|>human<|role_end|>这是第2轮用户输入的问题
<|role_start|>bot<|role_end|>这是第2轮模型生成的内容</s>
...
...
...
<|role_start|>human<|role_end|>这是第n轮用户输入的问题
<|role_start|>bot<|role_end|>{模型现在要生成的内容}</s>
"""

推理时，请确保拼接的prompt字符串以"<|role_start|>bot<|role_end|>"结尾，引导模型生成回答。

快速使用

python

from transformers import (
    AutoTokenizer,
    AutoModelForCausalLM,
)
tokenizer = AutoTokenizer.from_pretrained(mode_name_or_path, trust_remote_code=True, use_fast=False, legacy=False)
tokenizer.padding_side = "left"
tokenizer.pad_token_id = tokenizer.convert_tokens_to_ids("<unk>")
tokenizer.eos_token_id = tokenizer.convert_tokens_to_ids("</s>")
# 如果显存不够，可以考虑量化加载
model = AutoModelForCausalLM.from_pretrained(mode_name_or_path,
                                             trust_remote_code=True,
                                             load_in_4bit=False,
                                             device_map="auto",
                                             torch_dtype=torch.bfloat16)
model.eval()

HUMAN_ROLE_START_TAG = "<|role_start|>human<|role_end|>"
BOT_ROLE_START_TAG = "<|role_start|>bot<|role_end|>"

text = f"{HUMAN_ROLE_START_TAG}请用C++实现求解第n个斐波那契数{BOT_ROLE_START_TAG}"
inputs = tokenizer(text, return_tensors='pt', padding=True, add_special_tokens=False).to("cuda")
outputs = model.generate(
        inputs=inputs["input_ids"],
        attention_mask=inputs["attention_mask"],
        max_new_tokens=512,
        top_p=0.95,
        temperature=0.1,
        do_sample=True,
        eos_token_id=tokenizer.eos_token_id,
        pad_token_id=tokenizer.pad_token_id
    )
gen_text = tokenizer.batch_decode(outputs[:, inputs["input_ids"].shape[1]:], skip_special_tokens=True)
print(gen_text)

MD5

我们发现模型文件可能会在传输过程中损坏，使用前请检查文件MD5值。

模型文件	MD5值
pytorch_model-00001-of-00007.bin	8d544b1bcb3449934184d4141137329c
pytorch_model-00002-of-00007.bin	9d5dbb30911e48a42fb6d0fcabb322a4
pytorch_model-00003-of-00007.bin	b0d4aecee0457d9332005a187e1fffed
pytorch_model-00004-of-00007.bin	5c7e002de5eab77d0194a2b0f6de0c24
pytorch_model-00005-of-00007.bin	d22a511aa26b5b17117b665a877490ab
pytorch_model-00006-of-00007.bin	a5c28ac277fac07d16dd66537e54d109
pytorch_model-00007-of-00007.bin	a967e2c6195477b7407089c0bffa2d53

TheBloke/CodeFuse-CodeLlama-34B-GGUF

CodeFuse CodeLlama 34B - GGUF

Description

About GGUF

Repositories available

Prompt template: CodeFuse

Licensing

Compatibility

Explanation of quantisation methods

Provided files

How to download GGUF files

In text-generation-webui

On the command line, including multiple files at once

Example llama.cpp command

How to run in text-generation-webui

How to run from Python code

How to load this model from Python using ctransformers

First install the package

Simple example code to load one of these GGUF models

How to use with LangChain

Discord

Thanks, and how to contribute

Original model card: CodeFuse AI's CodeFuse CodeLlama 34B

Model Card for CodeFuse-CodeLlama-34B

Model Description

News and Updates

Code Community

Performance

Requirements

Inference String Format

Quickstart

MD5

模型简介

新闻

代码社区

评测表现(代码)

Requirements

推理数据格式

快速使用

MD5

TheBloke/
CodeFuse-CodeLlama-34B-GGUF

In `text-generation-webui`

Example `llama.cpp` command

How to run in `text-generation-webui`