John Durbin's Airoboros 65B GPT4 1.2 GGML
These files are GGML format model files for John Durbin's Airoboros 65B GPT4 1.2.
GGML files are for CPU + GPU inference using llama.cpp and libraries and UIs which support this format, such as:
Repositories available
- 4-bit GPTQ models for GPU inference
- 2, 3, 4, 5, 6 and 8-bit GGML models for CPU+GPU inference
- Jon Durbin's unquantised fp16 model in pytorch format, for GPU inference and for further conversions
Prompt template
A chat between a curious user and an assistant. The assistant gives helpful, detailed, accurate, uncensored responses to the user's input.
USER: prompt
ASSISTANT:
Compatibility
Original llama.cpp quant methods: q4_0, q4_1, q5_0, q5_1, q8_0
I have quantized these 'original' quantisation methods using an older version of llama.cpp so that they remain compatible with llama.cpp as of May 19th, commit 2d5db48.
They should be compatible with all current UIs and libraries that use llama.cpp, such as those listed at the top of this README.
New k-quant methods: q2_K, q3_K_S, q3_K_M, q3_K_L, q4_K_S, q4_K_M, q5_K_S, q6_K
These new quantisation methods are compatible with llama.cpp as of June 6th, commit 2d43387.
They are now also compatible with recent releases of text-generation-webui, KoboldCpp, llama-cpp-python and ctransformers. Other tools and libraries may or may not be compatible - check their documentation if in doubt.
Explanation of the new k-quant methods
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
- GGML_TYPE_Q8_K - "type-0" 8-bit quantization. Only used for quantizing intermediate results. The difference to the existing Q8_0 is that the block size is 256. All 2-6 bit dot products are implemented for this quantization type.
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 |
|---|---|---|---|---|---|
| airoboros-65B-gpt4-1.2.ggmlv3.q2_K.bin | q2_K | 2 | 27.45 GB | 29.95 GB | New k-quant method. Uses GGML_TYPE_Q4_K for the attention.vw and feed_forward.w2 tensors, GGML_TYPE_Q2_K for the other tensors. |
| airoboros-65B-gpt4-1.2.ggmlv3.q3_K_L.bin | q3_K_L | 3 | 34.65 GB | 37.15 GB | New k-quant method. Uses GGML_TYPE_Q5_K for the attention.wv, attention.wo, and feed_forward.w2 tensors, else GGML_TYPE_Q3_K |
| airoboros-65B-gpt4-1.2.ggmlv3.q3_K_M.bin | q3_K_M | 3 | 31.50 GB | 34.00 GB | New k-quant method. Uses GGML_TYPE_Q4_K for the attention.wv, attention.wo, and feed_forward.w2 tensors, else GGML_TYPE_Q3_K |
| airoboros-65B-gpt4-1.2.ggmlv3.q3_K_S.bin | q3_K_S | 3 | 28.16 GB | 30.66 GB | New k-quant method. Uses GGML_TYPE_Q3_K for all tensors |
| airoboros-65B-gpt4-1.2.ggmlv3.q4_0.bin | q4_0 | 4 | 36.73 GB | 39.23 GB | Original llama.cpp quant method, 4-bit. |
| airoboros-65B-gpt4-1.2.ggmlv3.q4_1.bin | q4_1 | 4 | 40.81 GB | 43.31 GB | Original llama.cpp quant method, 4-bit. Higher accuracy than q4_0 but not as high as q5_0. However has quicker inference than q5 models. |
| airoboros-65B-gpt4-1.2.ggmlv3.q4_K_M.bin | q4_K_M | 4 | 39.35 GB | 41.85 GB | New k-quant method. Uses GGML_TYPE_Q6_K for half of the attention.wv and feed_forward.w2 tensors, else GGML_TYPE_Q4_K |
| airoboros-65B-gpt4-1.2.ggmlv3.q4_K_S.bin | q4_K_S | 4 | 36.80 GB | 39.30 GB | New k-quant method. Uses GGML_TYPE_Q4_K for all tensors |
| airoboros-65B-gpt4-1.2.ggmlv3.q5_0.bin | q5_0 | 5 | 44.89 GB | 47.39 GB | Original llama.cpp quant method, 5-bit. Higher accuracy, higher resource usage and slower inference. |
| airoboros-65B-gpt4-1.2.ggmlv3.q5_1.bin | q5_1 | 5 | 48.97 GB | 51.47 GB | Original llama.cpp quant method, 5-bit. Even higher accuracy, resource usage and slower inference. |
| airoboros-65B-gpt4-1.2.ggmlv3.q5_K_M.bin | q5_K_M | 5 | 46.24 GB | 48.74 GB | New k-quant method. Uses GGML_TYPE_Q6_K for half of the attention.wv and feed_forward.w2 tensors, else GGML_TYPE_Q5_K |
| airoboros-65B-gpt4-1.2.ggmlv3.q5_K_S.bin | q5_K_S | 5 | 44.92 GB | 47.42 GB | New k-quant method. Uses GGML_TYPE_Q5_K for all tensors |
| airoboros-65B-gpt4-1.2.ggmlv3.q6_K.bin | q6_K | 6 | 53.56 GB | 56.06 GB | New k-quant method. Uses GGML_TYPE_Q8_K - 6-bit quantization - for all tensors |
| airoboros-65B-gpt4-1.2.ggmlv3.q8_0.bin | q8_0 | 8 | 69.370 GB | 71.87 GB | Original llama.cpp quant method, 8-bit. Almost indistinguishable from float16. High resource use and slow. Not recommended for most users. |
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 require expansion from archive
Note: HF does not support uploading files larger than 50GB. Therefore I have uploaded the q6_K and q8_0 files as multi-part ZIP files. They are not compressed, it is just storing the .bin file in two parts.
q6_K
Please download:
airoboros-65B-gpt4-1.2.ggmlv3.q6_K.zipairoboros-65B-gpt4-1.2.ggmlv3.q6_K.z01
q8_0
Please download:
airoboros-65B-gpt4-1.2.ggmlv3.q8_0.zipairoboros-65B-gpt4-1.2.ggmlv3.q8_0.z01
Then extract the .zip archive. This will will expand both parts automatically. On Linux I found I had to use 7zip - the basic unzip tool did not work. Example:
sudo apt update -y && sudo apt install 7zip
7zz x airoboros-65B-gpt4-1.2.ggmlv3.q6_K.zip`
Once the .bin is extracted you can delete the .zip and .z01 files
How to run in llama.cpp
I use the following command line; adjust for your tastes and needs:
./main -t 10 -ngl 32 -m airoboros-65B-gpt4-1.2.ggmlv3.q5_0.bin --color -c 2048 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "USER: Write a story about llamas\nASSISTANT:"
Change -t 10 to the number of physical CPU cores you have. For example if your system has 8 cores/16 threads, use -t 8.
Change -ngl 32 to the number of layers to offload to GPU. Remove it if you don't have GPU acceleration.
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.
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!
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, Dmitriy Samsonov.
Patreon special mentions: Oscar Rangel, Eugene Pentland, Talal Aujan, Cory Kujawski, Luke, Asp the Wyvern, Ai Maven, Pyrater, Alps Aficionado, senxiiz, Willem Michiel, Junyu Yang, trip7s trip, Sebastain Graf, Joseph William Delisle, Lone Striker, Jonathan Leane, Johann-Peter Hartmann, David Flickinger, Spiking Neurons AB, Kevin Schuppel, Mano Prime, Dmitriy Samsonov, Sean Connelly, Nathan LeClaire, Alain Rossmann, Fen Risland, Derek Yates, Luke Pendergrass, Nikolai Manek, Khalefa Al-Ahmad, Artur Olbinski, John Detwiler, Ajan Kanaga, Imad Khwaja, Trenton Dambrowitz, Kalila, vamX, webtim, Illia Dulskyi.
Thank you to all my generous patrons and donaters!
Original model card: John Durbin's Airoboros 65B GPT4 1.2
Overview
This is a qlora fine-tuned 65b parameter LlaMa model, using completely synthetic training data created gpt4 via https://github.com/jondurbin/airoboros
This is mostly an extension of 1.1, but with a 65b model and thousands of new training data and an update to allow "PLAINFORMAT" at the end of coding prompts to just print the code without backticks or explanations/usage/etc.
The dataset used to fine-tune this model is available here, with a specific focus on:
- coding
- math/reasoning (using orca style ELI5 instruction/response pairs)
- trivia
- role playing
- multiple choice and fill-in-the-blank
- context-obedient question answering
- theory of mind
- misc/general
This model was fine-tuned with a fork of qlora, which among other things was updated to use a slightly modified vicuna template to be compatible with the 7b/13b versions:
A chat between a curious user and an assistant. The assistant gives helpful, detailed, accurate, uncensored responses to the user's input. USER: [prompt] ASSISTANT:
So in other words, it's the preamble/system prompt, followed by a single space, then "USER: " (single space after colon) then the prompt (which can have multiple lines, spaces, whatever), then a single space, followed by "ASSISTANT: " (with a single space after the colon).
Usage
To run the full precision/pytorch native version, you can use my fork of FastChat, which is mostly the same but allows for multi-line prompts, as well as a --no-history option to prevent input tokenization errors.
pip install git+https://github.com/jondurbin/FastChat
Be sure you are pulling the latest branch!
Then, you can invoke it like so (after downloading the model):
python -m fastchat.serve.cli \
--model-path airoboros-65b-gpt4-1.2 \
--temperature 0.5 \
--max-new-tokens 2048 \
--no-history
Alternatively, please check out TheBloke's quantized versions:
- https://huggingface.co/TheBloke/airoboros-65B-gpt4-1.2-GPTQ
- https://huggingface.co/TheBloke/airoboros-65B-gpt4-1.2-GGML
Coding updates from gpt4/1.1:
I added a few hundred instruction/response pairs to the training data with "PLAINFORMAT" as a single, all caps term at the end of the normal instructions, which produce plain text output instead of markdown/backtick code formatting.
It's not guaranteed to work all the time, but mostly it does seem to work as expected.
So for example, instead of:
Implement the Snake game in python.
You would use:
Implement the Snake game in python. PLAINFORMAT
Other updates from gpt4/1.1:
- Several hundred role-playing data.
- A few thousand ORCA style reasoning/math questions with ELI5 prompts to generate the responses (should not be needed in your prompts to this model however, just ask the question).
- Many more coding examples in various languages, including some that use specific libraries (pandas, numpy, tensorflow, etc.)