Chat & support: my new Discord server

Want to contribute? TheBloke's Patreon page

Meta's Llama 2 70B Chat GPTQ

These files are GPTQ model files for Meta's Llama 2 70B Chat .

Multiple GPTQ parameter permutations are provided; see Provided Files below for details of the options provided, their parameters, and the software used to create them.

Many thanks to William Beauchamp from Chai for providing the hardware for these quantisations!

ExLlama support for 70B is here!

As of this commit , ExLlama has support for Llama 2 70B models.

Please make sure you update ExLlama to the latest version. If you are a text-generation-webui one-click user, you must first uninstall the ExLlama wheel, then clone ExLlama into text-generation-webui/repositories ; full instructions are below.

Now that we have ExLlama, that is the recommended loader to use for these models, as performance should be better than with AutoGPTQ and GPTQ-for-LLaMa, and you will be able to use the higher accuracy models, eg 128g + Act-Order.

Reminder: ExLlama does not support 3-bit models, so if you wish to try those quants, you will need to use AutoGPTQ or GPTQ-for-LLaMa.

AutoGPTQ and GPTQ-for-LLaMa requires latest version of Transformers

If you plan to use any of these quants with AutoGPTQ or GPTQ-for-LLaMa, your Transformers needs to be be using the latest Github code.

If you're using text-generation-webui and have updated to the latest version, this is done for you automatically.

If not, you can update it manually with:

pip3 install git+https://github.com/huggingface/transformers

Repositories available

• GPTQ models for GPU inference, with multiple quantisation parameter options.
• Original unquantised fp16 model in pytorch format, for GPU inference and for further conversions

Prompt template: Llama-2-Chat

[INST] <<SYS>>
You are a helpful, respectful and honest assistant. Always answer as helpfully as possible, while being safe.  Your answers should not include any harmful, unethical, racist, sexist, toxic, dangerous, or illegal content. Please ensure that your responses are socially unbiased and positive in nature. If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. If you don't know the answer to a question, please don't share false information.
<</SYS>>
{prompt}[/INST]

To continue a conversation:

[INST] <<SYS>>
You are a helpful, respectful and honest assistant. Always answer as helpfully as possible, while being safe.  Your answers should not include any harmful, unethical, racist, sexist, toxic, dangerous, or illegal content. Please ensure that your responses are socially unbiased and positive in nature. If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. If you don't know the answer to a question, please don't share false information.
<</SYS>>
{prompt}[/INST]{model_reply} 
[INST]{prompt}[/INST]

Provided files

Multiple quantisation parameters are provided, to allow you to choose the best one for your hardware and requirements.

Each separate quant is in a different branch. See below for instructions on fetching from different branches.

Branch	Bits	Group Size	Act Order (desc_act)	File Size	ExLlama Compatible?	Made With	Description
main	4	128	False	35.33 GB	True	AutoGPTQ	Most compatible option. Good inference speed in AutoGPTQ and GPTQ-for-LLaMa. Lower inference quality than other options.
gptq-4bit-32g-actorder_True	4	32	True	40.66 GB	True	AutoGPTQ	4-bit, with Act Order and group size. 32g gives highest possible inference quality, with maximum VRAM usage. Poor AutoGPTQ CUDA speed.
gptq-4bit-64g-actorder_True	4	64	True	37.99 GB	True	AutoGPTQ	4-bit, with Act Order and group size. 64g uses less VRAM than 32g, but with slightly lower accuracy. Poor AutoGPTQ CUDA speed.
gptq-4bit-128g-actorder_True	4	128	True	36.65 GB	True	AutoGPTQ	4-bit, with Act Order and group size. 128g uses even less VRAM, but with slightly lower accuracy. Poor AutoGPTQ CUDA speed.
gptq-3bit--1g-actorder_True	3	None	True	26.78 GB	False	AutoGPTQ	3-bit, with Act Order and no group size. Lowest possible VRAM requirements. May be lower quality than 3-bit 128g.
gptq-3bit-128g-actorder_False	3	128	False	28.03 GB	False	AutoGPTQ	3-bit, with group size 128g but no act-order. Slightly higher VRAM requirements than 3-bit None.
gptq-3bit-128g-actorder_True	3	128	True	28.03 GB	False	AutoGPTQ	3-bit, with group size 128g and act-order. Higher quality than 128g-False but poor AutoGPTQ CUDA speed.
gptq-3bit-64g-actorder_True	3	64	True	29.30 GB	False	AutoGPTQ	3-bit, with group size 64g and act-order. Highest quality 3-bit option. Poor AutoGPTQ CUDA speed.

How to download from branches

• In text-generation-webui, you can add :branch to the end of the download name, eg TheBloke/Llama-2-70B-chat-GPTQ:gptq-4bit-32g-actorder_True
• With Git, you can clone a branch with:

git clone --branch gptq-4bit-32g-actorder_True https://huggingface.co/TheBloke/Llama-2-70B-chat-GPTQ`

• In Python Transformers code, the branch is the revision parameter; see below.

How to easily download and use this model in text-generation-webui .

Please make sure you're using the latest version of text-generation-webui , which includes support for Llama 2 models.

It is strongly recommended to use the text-generation-webui one-click-installers unless you know how to make a manual install.

Use ExLlama (4-bit models only) - recommended option if you have enough VRAM for 4-bit

ExLlama has now been updated to support Llama 2 70B, but you will need to update ExLlama to the latest version.

By default text-generation-webui installs a pre-compiled wheel for ExLlama. Until text-generation-webui updates to reflect the ExLlama changes - which hopefully won't be long - you must uninstall that and then update ExLlama in the text-generation-webui/repositories directory. ExLlama will then compile its kernel on model load.

Note that this requires that your system is capable of compiling CUDA extensions, which may be an issue on Windows.

Instructions for Linux One Click Installer:

Change directory into the text-generation-webui main folder: cd /path/to/text-generation-webui

Activate the conda env of text-generation-webui:

source "installer_files/conda/etc/profile.d/conda.sh"
conda activate installer_files/env

Run: pip3 uninstall exllama

Run: cd repositories/exllama followed by git pull to update exllama.

Now launch text-generation-webui and follow the instructions below for downloading and running the model. ExLlama should build its kernel when the model first loads.

Downloading and running the model in text-generation-webui

Click the Model tab .

Under Download custom model or LoRA , enter TheBloke/Llama-2-70B-chat-GPTQ .

• To download from a specific branch, enter for example TheBloke/Llama-2-70B-chat-GPTQ:gptq-4bit-32g-actorder_True
• see Provided Files above for the list of branches for each option.

Click Download .

The model will start downloading. Once it's finished it will say "Done"

Set Loader to ExLlama if you plan to use a 4-bit file, or else choose AutoGPTQ or GPTQ-for-LLaMA.

• If you use AutoGPTQ, make sure "No inject fused attention" is ticked

In the top left, click the refresh icon next to Model .

In the Model dropdown, choose the model you just downloaded: TheBloke/Llama-2-70B-chat-GPTQ

The model will automatically load, and is now ready for use!

Then click Save settings for this model followed by Reload the Model in the top right to make sure your settings are persisted.

Once you're ready, click the Text Generation tab and enter a prompt to get started!

How to use this GPTQ model from Python code

First make sure you have AutoGPTQ installed:

GITHUB_ACTIONS=true pip3 install auto-gptq

You also need the latest Transformers code from Github:

pip3 install git+https://github.com/huggingface/transformers

You must set inject_fused_attention=False as shown below.

Then try the following example code:

from transformers import AutoTokenizer, pipeline, logging
from auto_gptq import AutoGPTQForCausalLM, BaseQuantizeConfig

model_name_or_path = "TheBloke/Llama-2-70B-chat-GPTQ"
model_basename = "gptq_model-4bit--1g"

use_triton = False

tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)

model = AutoGPTQForCausalLM.from_quantized(model_name_or_path,
        model_basename=model_basename,
        inject_fused_attention=False, # Required for Llama 2 70B model at this time.
        use_safetensors=True,
        trust_remote_code=False,
        device="cuda:0",
        use_triton=use_triton,
        quantize_config=None)

"""
To download from a specific branch, use the revision parameter, as in this example:

model = AutoGPTQForCausalLM.from_quantized(model_name_or_path,
        revision="gptq-4bit-32g-actorder_True",
        model_basename=model_basename,
        inject_fused_attention=False, # Required for Llama 2 70B model at this time.
        use_safetensors=True,
        trust_remote_code=False,
        device="cuda:0",
        quantize_config=None)
"""

prompt = "Tell me about AI"
prompt_template=f'''SYSTEM: You are a helpful, respectful and honest assistant. Always answer as helpfully as possible, while being safe.  Your answers should not include any harmful, unethical, racist, sexist, toxic, dangerous, or illegal content. Please ensure that your responses are socially unbiased and positive in nature. If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. If you don't know the answer to a question, please don't share false information.
USER: {prompt}
ASSISTANT:
'''

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

input_ids = tokenizer(prompt_template, return_tensors='pt').input_ids.cuda()
output = model.generate(inputs=input_ids, temperature=0.7, max_new_tokens=512)
print(tokenizer.decode(output[0]))

# Inference can also be done using transformers' pipeline

# Prevent printing spurious transformers error when using pipeline with AutoGPTQ
logging.set_verbosity(logging.CRITICAL)

print("*** Pipeline:")
pipe = pipeline(
    "text-generation",
    model=model,
    tokenizer=tokenizer,
    max_new_tokens=512,
    temperature=0.7,
    top_p=0.95,
    repetition_penalty=1.15
)

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

Compatibility

The files provided will work with AutoGPTQ (CUDA and Triton modes), GPTQ-for-LLaMa (only CUDA has been tested), and Occ4m's GPTQ-for-LLaMa fork.

ExLlama is now compatible with Llama 2 70B models, as of this commit .

Please see the Provided Files table above for per-file compatibility.

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!

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 : Luke from CarbonQuill, Aemon Algiz.

Patreon special mentions : Space Cruiser, Nikolai Manek, Sam, Chris McCloskey, Rishabh Srivastava, Kalila, Spiking Neurons AB, Khalefa Al-Ahmad, WelcomeToTheClub, Chadd, Lone Striker, Viktor Bowallius, Edmond Seymore, Ai Maven, Chris Smitley, Dave, Alexandros Triantafyllidis, Luke @flexchar, Elle, ya boyyy, Talal Aujan, Alex , Jonathan Leane, Deep Realms, Randy H, subjectnull, Preetika Verma, Joseph William Delisle, Michael Levine, chris gileta, K, Oscar Rangel, LangChain4j, Trenton Dambrowitz, Eugene Pentland, Johann-Peter Hartmann, Femi Adebogun, Illia Dulskyi, senxiiz, Daniel P. Andersen, Sean Connelly, Artur Olbinski, RoA, Mano Prime, Derek Yates, Raven Klaugh, David Flickinger, Willem Michiel, Pieter, Willian Hasse, vamX, Luke Pendergrass, webtim, Ghost , Rainer Wilmers, Nathan LeClaire, Will Dee, Cory Kujawski, John Detwiler, Fred von Graf, biorpg, Iucharbius , Imad Khwaja, Pierre Kircher, terasurfer , Asp the Wyvern, John Villwock, theTransient, zynix , Gabriel Tamborski, Fen Risland, Gabriel Puliatti, Matthew Berman, Pyrater, SuperWojo, Stephen Murray, Karl Bernard, Ajan Kanaga, Greatston Gnanesh, Junyu Yang.

Thank you to all my generous patrons and donaters!

Original model card: Meta's Llama 2 70B Chat

Llama 2

Llama 2 is a collection of pretrained and fine-tuned generative text models ranging in scale from 7 billion to 70 billion parameters. This is the repository for the 70B fine-tuned model, optimized for dialogue use cases and converted for the Hugging Face Transformers format. Links to other models can be found in the index at the bottom.

Model Details

Note: Use of this model is governed by the Meta license. In order to download the model weights and tokenizer, please visit the website and accept our License before requesting access here.

Meta developed and publicly released the Llama 2 family of large language models (LLMs), a collection of pretrained and fine-tuned generative text models ranging in scale from 7 billion to 70 billion parameters. Our fine-tuned LLMs, called Llama-2-Chat, are optimized for dialogue use cases. Llama-2-Chat models outperform open-source chat models on most benchmarks we tested, and in our human evaluations for helpfulness and safety, are on par with some popular closed-source models like ChatGPT and PaLM.

Model Developers Meta

Variations Llama 2 comes in a range of parameter sizes — 7B, 13B, and 70B — as well as pretrained and fine-tuned variations.

Input Models input text only.

Output Models generate text only.

Model Architecture Llama 2 is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align to human preferences for helpfulness and safety.

Training Data	Params	Content Length	GQA	Tokens	LR
Llama 2	A new mix of publicly available online data	7B	4k	✗	2.0T	3.0 x 10 -4
Llama 2	A new mix of publicly available online data	13B	4k	✗	2.0T	3.0 x 10 -4
Llama 2	A new mix of publicly available online data	70B	4k	✔	2.0T	1.5 x 10 -4

Llama 2 family of models. Token counts refer to pretraining data only. All models are trained with a global batch-size of 4M tokens. Bigger models - 70B -- use Grouped-Query Attention (GQA) for improved inference scalability.

Model Dates Llama 2 was trained between January 2023 and July 2023.

Status This is a static model trained on an offline dataset. Future versions of the tuned models will be released as we improve model safety with community feedback.

License A custom commercial license is available at: https://ai.meta.com/resources/models-and-libraries/llama-downloads/

Intended Use

Intended Use Cases Llama 2 is intended for commercial and research use in English. Tuned models are intended for assistant-like chat, whereas pretrained models can be adapted for a variety of natural language generation tasks.

Out-of-scope Uses Use in any manner that violates applicable laws or regulations (including trade compliance laws).Use in languages other than English. Use in any other way that is prohibited by the Acceptable Use Policy and Licensing Agreement for Llama 2.

Hardware and Software

Training Factors We used custom training libraries, Meta's Research Super Cluster, and production clusters for pretraining. Fine-tuning, annotation, and evaluation were also performed on third-party cloud compute.

Carbon Footprint Pretraining utilized a cumulative 3.3M GPU hours of computation on hardware of type A100-80GB (TDP of 350-400W). Estimated total emissions were 539 tCO2eq, 100% of which were offset by Meta’s sustainability program.

Time (GPU hours)	Power Consumption (W)	Carbon Emitted(tCO 2 eq)
Llama 2 7B	184320	400	31.22
Llama 2 13B	368640	400	62.44
Llama 2 70B	1720320	400	291.42
Total	3311616	539.00

CO 2 emissions during pretraining. Time: total GPU time required for training each model. Power Consumption: peak power capacity per GPU device for the GPUs used adjusted for power usage efficiency. 100% of the emissions are directly offset by Meta's sustainability program, and because we are openly releasing these models, the pretraining costs do not need to be incurred by others.

Training Data

Overview Llama 2 was pretrained on 2 trillion tokens of data from publicly available sources. The fine-tuning data includes publicly available instruction datasets, as well as over one million new human-annotated examples. Neither the pretraining nor the fine-tuning datasets include Meta user data.

Data Freshness The pretraining data has a cutoff of September 2022, but some tuning data is more recent, up to July 2023.

Evaluation Results

In this section, we report the results for the Llama 1 and Llama 2 models on standard academic benchmarks.For all the evaluations, we use our internal evaluations library.

Model	Size	Code	Commonsense Reasoning	World Knowledge	Reading Comprehension	Math	MMLU	BBH	AGI Eval
Llama 1	7B	14.1	60.8	46.2	58.5	6.95	35.1	30.3	23.9
Llama 1	13B	18.9	66.1	52.6	62.3	10.9	46.9	37.0	33.9
Llama 1	33B	26.0	70.0	58.4	67.6	21.4	57.8	39.8	41.7
Llama 1	65B	30.7	70.7	60.5	68.6	30.8	63.4	43.5	47.6
Llama 2	7B	16.8	63.9	48.9	61.3	14.6	45.3	32.6	29.3
Llama 2	13B	24.5	66.9	55.4	65.8	28.7	54.8	39.4	39.1
Llama 2	70B	37.5	71.9	63.6	69.4	35.2	68.9	51.2	54.2

Overall performance on grouped academic benchmarks. Code: We report the average pass@1 scores of our models on HumanEval and MBPP. Commonsense Reasoning: We report the average of PIQA, SIQA, HellaSwag, WinoGrande, ARC easy and challenge, OpenBookQA, and CommonsenseQA. We report 7-shot results for CommonSenseQA and 0-shot results for all other benchmarks. World Knowledge: We evaluate the 5-shot performance on NaturalQuestions and TriviaQA and report the average. Reading Comprehension: For reading comprehension, we report the 0-shot average on SQuAD, QuAC, and BoolQ. MATH: We report the average of the GSM8K (8 shot) and MATH (4 shot) benchmarks at top 1.

TruthfulQA	Toxigen
Llama 1	7B	27.42	23.00
Llama 1	13B	41.74	23.08
Llama 1	33B	44.19	22.57
Llama 1	65B	48.71	21.77
Llama 2	7B	33.29	21.25
Llama 2	13B	41.86	26.10
Llama 2	70B	50.18	24.60

Evaluation of pretrained LLMs on automatic safety benchmarks. For TruthfulQA, we present the percentage of generations that are both truthful and informative (the higher the better). For ToxiGen, we present the percentage of toxic generations (the smaller the better).

TruthfulQA	Toxigen
Llama-2-Chat	7B	57.04	0.00
Llama-2-Chat	13B	62.18	0.00
Llama-2-Chat	70B	64.14	0.01

Evaluation of fine-tuned LLMs on different safety datasets. Same metric definitions as above.

Ethical Considerations and Limitations

Llama 2 is a new technology that carries risks with use. Testing conducted to date has been in English, and has not covered, nor could it cover all scenarios. For these reasons, as with all LLMs, Llama 2’s potential outputs cannot be predicted in advance, and the model may in some instances produce inaccurate, biased or other objectionable responses to user prompts. Therefore, before deploying any applications of Llama 2, developers should perform safety testing and tuning tailored to their specific applications of the model.

Please see the Responsible Use Guide available at https://ai.meta.com/llama/responsible-use-guide/

Reporting Issues

Please report any software “bug,” or other problems with the models through one of the following means:

• Reporting issues with the model: github.com/facebookresearch/llama
• Reporting problematic content generated by the model: developers.facebook.com/llama_output_feedback
• Reporting bugs and security concerns: facebook.com/whitehat/info

Llama Model Index

Model	Llama2	Llama2-hf	Llama2-chat	Llama2-chat-hf
7B	Link	Link	Link	Link
13B	Link	Link	Link	Link
70B	Link	Link	Link	Link