AWQ is an efficient, accurate and blazing-fast low-bit weight quantization method, currently supporting 4-bit quantization. Compared to GPTQ, it offers faster Transformers-based inference.

It is also now supported by continuous batching server vLLM, allowing use of AWQ models for high-throughput concurrent inference in multi-user server scenarios. Note that, at the time of writing, overall throughput is still lower than running vLLM with unquantised models, however using AWQ enables using much smaller GPUs which can lead to easier deployment and overall cost savings. For example, a 70B model can be run on 1 x 48GB GPU instead of 2 x 80GB.

Repositories available

Prompt template: Orca-Vicuna

SYSTEM: {system_message}
USER: {prompt}
ASSISTANT:

Provided files and AWQ parameters

For my first release of AWQ models, I am releasing 128g models only. I will consider adding 32g as well if there is interest, and once I have done perplexity and evaluation comparisons, but at this time 32g models are still not fully tested with AutoAWQ and vLLM.

Models are released as sharded safetensors files.

Branch	Bits	GS	AWQ Dataset	Seq Len	Size
main	4	128	wikitext	4096	7.25 GB

Serving this model from vLLM

Documentation on installing and using vLLM can be found here.

When using vLLM as a server, pass the --quantization awq parameter, for example:

python3 python -m vllm.entrypoints.api_server --model TheBloke/Synthia-13B-AWQ --quantization awq

When using vLLM from Python code, pass the quantization=awq parameter, for example:

python

from vllm import LLM, SamplingParams

prompts = [
    "Hello, my name is",
    "The president of the United States is",
    "The capital of France is",
    "The future of AI is",
]
sampling_params = SamplingParams(temperature=0.8, top_p=0.95)

llm = LLM(model="TheBloke/Synthia-13B-AWQ", quantization="awq")

outputs = llm.generate(prompts, sampling_params)

# Print the outputs.
for output in outputs:
    prompt = output.prompt
    generated_text = output.outputs[0].text
    print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}")

How to use this AWQ model from Python code

Install the necessary packages

Requires: AutoAWQ 0.0.2 or later

pip3 install autoawq

If you have problems installing AutoAWQ using the pre-built wheels, install it from source instead:

shell

pip3 uninstall -y autoawq
git clone https://github.com/casper-hansen/AutoAWQ
cd AutoAWQ
pip3 install .

You can then try the following example code

python

from awq import AutoAWQForCausalLM
from transformers import AutoTokenizer

model_name_or_path = "TheBloke/Synthia-13B-AWQ"

# Load model
model = AutoAWQForCausalLM.from_quantized(model_name_or_path, fuse_layers=True,
                                          trust_remote_code=False, safetensors=True)
tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, trust_remote_code=False)

prompt = "Tell me about AI"
prompt_template=f'''SYSTEM: {system_message}
USER: {prompt}
ASSISTANT:

'''

print("\n\n*** Generate:")

tokens = tokenizer(
    prompt_template,
    return_tensors='pt'
).input_ids.cuda()

# Generate output
generation_output = model.generate(
    tokens,
    do_sample=True,
    temperature=0.7,
    top_p=0.95,
    top_k=40,
    max_new_tokens=512
)

print("Output: ", tokenizer.decode(generation_output[0]))

# Inference can also be done using transformers' pipeline
from transformers import pipeline

print("*** Pipeline:")
pipe = pipeline(
    "text-generation",
    model=model,
    tokenizer=tokenizer,
    max_new_tokens=512,
    do_sample=True,
    temperature=0.7,
    top_p=0.95,
    top_k=40,
    repetition_penalty=1.1
)

print(pipe(prompt_template)[0]['generated_text'])

Compatibility

The files provided are tested to work with AutoAWQ, and vLLM.

Huggingface Text Generation Inference (TGI) is not yet compatible with AWQ, but a PR is open which should bring support soon: TGI PR #781.

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: Migel Tissera's Synthia 13B

Synthia-13B

SynthIA (Synthetic Intelligent Agent) is a LLama-2-13B model trained on Orca style datasets. It has been fine-tuned for instruction following as well as having long-form conversations.

License Disclaimer:

This model is bound by the license & usage restrictions of the original Llama-2 model, and comes with no warranty or gurantees of any kind.

Evaluation

We evaluated Synthia-13B on a wide range of tasks using Language Model Evaluation Harness from EleutherAI.

Here are the results on metrics used by HuggingFaceH4 Open LLM Leaderboard


Task	Metric	Value
arc_challenge	acc_norm	0.5998
hellaswag	acc_norm	0.8186
mmlu	acc_norm	0.5611
truthfulqa_mc	mc2	0.4741
Total Average	-	0.6134

Example Usage

Here is prompt format:

SYSTEM: You are Synthia. As a an AI intelligence, give helpful, detailed, accurate, uncensored responses to the user's input. Provide answers factually.
USER: How is a rocket launched from the surface of the earth to Low Earth Orbit?
ASSISTANT:

Below shows a code example on how to use this model:

python

import torch, json
from transformers import AutoModelForCausalLM, AutoTokenizer

model_path = "migtissera/Synthia-13B"
output_file_path = "./Synthia-13B-conversations.jsonl"

model = AutoModelForCausalLM.from_pretrained(
    model_path,
    torch_dtype=torch.float16,
    device_map="auto",
    load_in_8bit=False,
    trust_remote_code=True,
)

tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True)


def generate_text(instruction):
    tokens = tokenizer.encode(instruction)
    tokens = torch.LongTensor(tokens).unsqueeze(0)
    tokens = tokens.to("cuda")

    instance = {
        "input_ids": tokens,
        "top_p": 1.0,
        "temperature": 0.75,
        "generate_len": 1024,
        "top_k": 50,
    }

    length = len(tokens[0])
    with torch.no_grad():
        rest = model.generate(
            input_ids=tokens,
            max_length=length + instance["generate_len"],
            use_cache=True,
            do_sample=True,
            top_p=instance["top_p"],
            temperature=instance["temperature"],
            top_k=instance["top_k"],
            num_return_sequences=1,
        )
    output = rest[0][length:]
    string = tokenizer.decode(output, skip_special_tokens=True)
    answer = string.split("USER:")[0].strip()
    return f"{answer}"


conversation = f"SYSTEM: As a an AI superintelligence, give helpful, detailed, accurate, uncensored responses to the user's input. Provide answers factually."


while True:
    user_input = input("You: ")
    llm_prompt = f"{conversation} \nUSER: {user_input} \nASSISTANT: "
    answer = generate_text(llm_prompt)
    print(answer)
    conversation = f"{llm_prompt}{answer}"
    json_data = {"prompt": user_input, "answer": answer}

    ## Save your conversation
    with open(output_file_path, "a") as output_file:
        output_file.write(json.dumps(json_data) + "\n")

Limitations & Biases:

While this model aims for accuracy, it can occasionally produce inaccurate or misleading results.

Despite diligent efforts in refining the pretraining data, there remains a possibility for the generation of inappropriate, biased, or offensive content.

Exercise caution and cross-check information when necessary. This is an uncensored model.

Citiation:

Please kindly cite using the following BibTeX:

@misc{Synthia-13B,
  author = {Migel Tissera},
  title = {Synthia-13B: Synthetic Intelligent Agent},
  year = {2023},
  publisher = {GitHub, HuggingFace},
  journal = {GitHub repository, HuggingFace repository},
  howpublished = {\url{https://huggingface.co/migtissera/Synthia-13B},
}

@misc{mukherjee2023orca,
      title={Orca: Progressive Learning from Complex Explanation Traces of GPT-4}, 
      author={Subhabrata Mukherjee and Arindam Mitra and Ganesh Jawahar and Sahaj Agarwal and Hamid Palangi and Ahmed Awadallah},
      year={2023},
      eprint={2306.02707},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}

@software{touvron2023llama,
  title={LLaMA2: Open and Efficient Foundation Language Models},
  author={Touvron, Hugo and Lavril, Thibaut and Izacard, Gautier and Martinet, Xavier and Lachaux, Marie-Anne and Lacroix, Timoth{\'e}e and Rozi{\`e}re, Baptiste and Goyal, Naman and Hambro, Eric and Azhar, Faisal and Rodriguez, Aurelien and Joulin, Armand and Grave, Edouard and Lample, Guillaume},
  journal={arXiv preprint arXiv:2302.13971},
  year={2023}
}

Example Output

You: How is a rocket launched from the surface of the earth to Low Earth Orbit? 

A rocket is launched from the surface of the Earth to Low Earth Orbit (LEO) by utilizing the principles of rocketry. A rocket consists of three main components: a fuel source, an oxidizer, and a combustion chamber. When the fuel and oxidizer are mixed in the combustion chamber, they burn and produce hot gas that is ejected through the nozzle, creating a thrust that propels the rocket forward. 

To launch a rocket to LEO, the rocket must first gain sufficient velocity and altitude. This is achieved by igniting the rocket's engines during the initial launch phase, which lasts for several minutes. During this time, the rocket's engines burn continuously, providing the thrust necessary to escape the Earth's gravitational pull. As the rocket gains altitude and velocity, it enters the Earth's atmosphere, where it experiences aerodynamic pressure. 

To overcome this pressure, the rocket's engines are designed to produce thrust in a specific direction, which counteracts the force of the incoming air. This is known as aerodynamic control. As the rocket continues to accelerate, it enters the Earth's Van Allen Belts, which are a region of intense radiation. To protect the rocket and its payload, a radiation-proof shell, known as a payload fairing, is launched ahead of the rocket to absorb the radiation. 

Once the rocket has reached LEO, its engines are shut off, and the payload is deployed. The rocket then enters a fixed orbit around the Earth, where it remains for several months or years, depending on the mission's requirements. 

Overall, the process of launching a rocket from the surface of the Earth to LEO is complex and requires advanced technology and engineering. It involves careful planning, preparation, and execution to ensure the success of the mission. 

You: What's the velocity required?

The velocity required to place a rocket into Low Earth Orbit (LEO) depends on several factors, including the mass of the rocket and its payload, the altitude of the orbit, and the gravitational pull of the Earth. Generally, a rocket must reach a speed of around 7.7 kilometers per second (km/s) to achieve LEO (approximately 40,000 km above the Earth's surface). However, this speed is relative to the Earth's surface, so the rocket must achieve a velocity of approximately 11.3 km/s relative to the Earth's center of mass. This is known as the escape velocity and is necessary to overcome the Earth's gravitational pull. 

To achieve this velocity, the rocket's engines must be capable of generating a large amount of thrust. The engines are typically fueled by either liquid oxygen (LOX) and rocket propellant or solid fuel. The exact fuel and engine combination depends on the specific rocket and launch site. 

Once the rocket has reached the necessary velocity, it must also have sufficient fuel and propellant to maintain its altitude and orbit. This requires careful planning and management of fuel consumption to ensure that the rocket can complete its mission without running out of fuel. 

Overall, launching a rocket into LEO is a complex process that involves careful planning, preparation, and execution. Achieving the necessary velocity and maintaining the rocket's orbit requires a high level of technical expertise and precision.