TheBloke's LLM work is generously supported by a grant from andreessen horowitz (a16z)
Yarn Llama 2 70B 32K - GGUF
- Model creator: NousResearch
- Original model: Yarn Llama 2 70B 32K
Description
This repo contains GGUF format model files for NousResearch's Yarn Llama 2 70B 32K.
These files were quantised using hardware kindly provided by Massed Compute.
About GGUF
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.
Here is an incomplete 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
- AWQ model(s) for GPU inference.
- GPTQ models for GPU inference, with multiple quantisation parameter options.
- 2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference
- NousResearch's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions
Prompt template: None
{prompt}
Licensing
The creator of the source model has listed its license as apache-2.0, 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: NousResearch's Yarn Llama 2 70B 32K.
Compatibility
These quantised GGUFv2 files are compatible with llama.cpp from August 27th onwards, as of commit d0cee0d
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 |
|---|---|---|---|---|---|
| yarn-llama-2-70b-32k.Q2_K.gguf | Q2_K | 2 | 29.28 GB | 31.78 GB | smallest, significant quality loss - not recommended for most purposes |
| yarn-llama-2-70b-32k.Q3_K_S.gguf | Q3_K_S | 3 | 29.92 GB | 32.42 GB | very small, high quality loss |
| yarn-llama-2-70b-32k.Q3_K_M.gguf | Q3_K_M | 3 | 33.19 GB | 35.69 GB | very small, high quality loss |
| yarn-llama-2-70b-32k.Q3_K_L.gguf | Q3_K_L | 3 | 36.15 GB | 38.65 GB | small, substantial quality loss |
| yarn-llama-2-70b-32k.Q4_0.gguf | Q4_0 | 4 | 38.87 GB | 41.37 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
| yarn-llama-2-70b-32k.Q4_K_S.gguf | Q4_K_S | 4 | 39.07 GB | 41.57 GB | small, greater quality loss |
| yarn-llama-2-70b-32k.Q4_K_M.gguf | Q4_K_M | 4 | 41.42 GB | 43.92 GB | medium, balanced quality - recommended |
| yarn-llama-2-70b-32k.Q5_0.gguf | Q5_0 | 5 | 47.46 GB | 49.96 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
| yarn-llama-2-70b-32k.Q5_K_S.gguf | Q5_K_S | 5 | 47.46 GB | 49.96 GB | large, low quality loss - recommended |
| yarn-llama-2-70b-32k.Q5_K_M.gguf | Q5_K_M | 5 | 48.75 GB | 51.25 GB | large, very low quality loss - recommended |
| yarn-llama-2-70b-32k.Q6_K.gguf | Q6_K | 6 | 56.59 GB | 59.09 GB | very large, extremely low quality loss |
| yarn-llama-2-70b-32k.Q8_0.gguf | Q8_0 | 8 | 73.29 GB | 75.79 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.
Q6_K and Q8_0 files are split and require joining
Note: HF does not support uploading files larger than 50GB. Therefore I have uploaded the Q6_K and Q8_0 files as split files.
Click for instructions regarding Q6_K and Q8_0 files
q6_K
Please download:
yarn-llama-2-70b-32k.Q6_K.gguf-split-ayarn-llama-2-70b-32k.Q6_K.gguf-split-b
q8_0
Please download:
yarn-llama-2-70b-32k.Q8_0.gguf-split-ayarn-llama-2-70b-32k.Q8_0.gguf-split-b
To join the files, do the following:
Linux and macOS:
cat yarn-llama-2-70b-32k.Q6_K.gguf-split-* > yarn-llama-2-70b-32k.Q6_K.gguf && rm yarn-llama-2-70b-32k.Q6_K.gguf-split-*
cat yarn-llama-2-70b-32k.Q8_0.gguf-split-* > yarn-llama-2-70b-32k.Q8_0.gguf && rm yarn-llama-2-70b-32k.Q8_0.gguf-split-*
Windows command line:
COPY /B yarn-llama-2-70b-32k.Q6_K.gguf-split-a + yarn-llama-2-70b-32k.Q6_K.gguf-split-b yarn-llama-2-70b-32k.Q6_K.gguf
del yarn-llama-2-70b-32k.Q6_K.gguf-split-a yarn-llama-2-70b-32k.Q6_K.gguf-split-b
COPY /B yarn-llama-2-70b-32k.Q8_0.gguf-split-a + yarn-llama-2-70b-32k.Q8_0.gguf-split-b yarn-llama-2-70b-32k.Q8_0.gguf
del yarn-llama-2-70b-32k.Q8_0.gguf-split-a yarn-llama-2-70b-32k.Q8_0.gguf-split-b
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/Yarn-Llama-2-70B-32k-GGUF and below it, a specific filename to download, such as: yarn-llama-2-70b-32k.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
Then you can download any individual model file to the current directory, at high speed, with a command like this:
huggingface-cli download TheBloke/Yarn-Llama-2-70B-32k-GGUF yarn-llama-2-70b-32k.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/Yarn-Llama-2-70B-32k-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:
HF_HUB_ENABLE_HF_TRANSFER=1 huggingface-cli download TheBloke/Yarn-Llama-2-70B-32k-GGUF yarn-llama-2-70b-32k.Q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
Windows Command Line users: You can set the environment variable by running set HF_HUB_ENABLE_HF_TRANSFER=1 before the download command.
Example llama.cpp command
Make sure you are using llama.cpp from commit d0cee0d or later.
./main -ngl 32 -m yarn-llama-2-70b-32k.Q4_K_M.gguf --color -c 4096 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "{prompt}"
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 can be found in the text-generation-webui documentation, here: text-generation-webui/docs/04 ‐ Model Tab.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 in Python code, using ctransformers
First install the package
Run one of the following commands, according to your system:
# Base ctransformers with no GPU acceleration
pip install ctransformers
# Or with CUDA GPU acceleration
pip install ctransformers[cuda]
# Or with AMD ROCm GPU acceleration (Linux only)
CT_HIPBLAS=1 pip install ctransformers --no-binary ctransformers
# Or with Metal GPU acceleration for macOS systems only
CT_METAL=1 pip install ctransformers --no-binary ctransformers
Simple ctransformers example code
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/Yarn-Llama-2-70B-32k-GGUF", model_file="yarn-llama-2-70b-32k.Q4_K_M.gguf", model_type="llama", gpu_layers=50)
print(llm("AI is going to"))
How to use with LangChain
Here are guides on using llama-cpp-python and ctransformers with LangChain:
Discord
For further support, and discussions on these models and AI in general, join us at:
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: Brandon Frisco, LangChain4j, Spiking Neurons AB, transmissions 11, Joseph William Delisle, Nitin Borwankar, Willem Michiel, Michael Dempsey, vamX, Jeffrey Morgan, zynix, jjj, Omer Bin Jawed, Sean Connelly, jinyuan sun, Jeromy Smith, Shadi, Pawan Osman, Chadd, Elijah Stavena, Illia Dulskyi, Sebastain Graf, Stephen Murray, terasurfer, Edmond Seymore, Celu Ramasamy, Mandus, Alex, biorpg, Ajan Kanaga, Clay Pascal, Raven Klaugh, 阿明, K, ya boyyy, usrbinkat, Alicia Loh, John Villwock, ReadyPlayerEmma, Chris Smitley, Cap'n Zoog, fincy, GodLy, S_X, sidney chen, Cory Kujawski, OG, Mano Prime, AzureBlack, Pieter, Kalila, Spencer Kim, Tom X Nguyen, Stanislav Ovsiannikov, Michael Levine, Andrey, Trailburnt, Vadim, Enrico Ros, Talal Aujan, Brandon Phillips, Jack West, Eugene Pentland, Michael Davis, Will Dee, webtim, Jonathan Leane, Alps Aficionado, Rooh Singh, Tiffany J. Kim, theTransient, Luke @flexchar, Elle, Caitlyn Gatomon, Ari Malik, subjectnull, Johann-Peter Hartmann, Trenton Dambrowitz, Imad Khwaja, Asp the Wyvern, Emad Mostaque, Rainer Wilmers, Alexandros Triantafyllidis, Nicholas, Pedro Madruga, SuperWojo, Harry Royden McLaughlin, James Bentley, Olakabola, David Ziegler, Ai Maven, Jeff Scroggin, Nikolai Manek, Deo Leter, Matthew Berman, Fen Risland, Ken Nordquist, Manuel Alberto Morcote, Luke Pendergrass, TL, Fred von Graf, Randy H, Dan Guido, NimbleBox.ai, Vitor Caleffi, Gabriel Tamborski, knownsqashed, Lone Striker, Erik Bjäreholt, John Detwiler, Leonard Tan, Iucharbius
Thank you to all my generous patrons and donaters!
And thank you again to a16z for their generous grant.
Original model card: NousResearch's Yarn Llama 2 70B 32K
Model Card: Yarn-Llama-2-70b-32k
The authors would like to thank LAION AI for their support of compute for this model. It was trained on the JUWELS supercomputer.
Model Description
Nous-Yarn-Llama-2-70b-32k is a state-of-the-art language model for long context, further pretrained on long context data for 400 steps using the YaRN extension method. It is an extension of Llama-2-70b-hf and supports a 32k token context window.
To use, pass trust_remote_code=True when loading the model, for example
model = AutoModelForCausalLM.from_pretrained("NousResearch/Yarn-Llama-2-70b-32k",
use_flash_attention_2=True,
torch_dtype=torch.bfloat16,
device_map="auto",
trust_remote_code=True)
In addition you will need to use the latest version of transformers (until 4.35 comes out)
pip install git+https://github.com/huggingface/transformers
Benchmarks
Long context benchmarks:
| Model | Context Window | 1k PPL | 2k PPL | 4k PPL | 8k PPL | 16k PPL | 32k PPL |
|---|---|---|---|---|---|---|---|
| Llama-2-70b-hf | 4k | 3.71 | 3.27 | 2.96 | - | - | - |
| Yarn-Llama-2-70b-32k | 32k | 3.61 | 3.22 | 2.91 | 2.82 | 2.45 | 2.23 |
Short context benchmarks showing that quality degradation is minimal:
| Model | Context Window | ARC-c | MMLU | Truthful QA |
|---|---|---|---|---|
| Llama-2-70b-hf | 4k | 67.32 | 69.83 | 44.92 |
| Yarn-Llama-2-70b-32k | 32k | 67.41 | 68.84 | 46.14 |
Collaborators
- bloc97: Methods, paper and evals
- @theemozilla: Methods, paper, model training, and evals
- @EnricoShippole: Model training
- honglu2875: Paper and evals
- Downloads last month
- 246
Model tree for TheBloke/Yarn-Llama-2-70B-32k-GGUF
Base model
NousResearch/Yarn-Llama-2-70b-32k