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.
GPT4All, a free and open source local running GUI, supporting Windows, Linux and macOS with full GPU accel.
LM Studio, an easy-to-use and powerful local GUI for Windows and macOS (Silicon), with GPU acceleration. Linux available, in beta as of 27/11/2023.
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.
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.
ctransformers, a Python library with GPU accel, LangChain support, and OpenAI-compatible AI server. Note, as of time of writing (November 27th 2023), ctransformers has not been updated in a long time and does not support many recent models.

Repositories available

Prompt template: Alpaca

Below is an instruction that describes a task. Write a response that appropriately completes the request.

### Instruction:
{prompt}

### Response:

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
wizardcoder-33b-v1.1.Q2_K.gguf	Q2_K	2	14.03 GB	16.53 GB	smallest, significant quality loss - not recommended for most purposes
wizardcoder-33b-v1.1.Q3_K_S.gguf	Q3_K_S	3	14.42 GB	16.92 GB	very small, high quality loss
wizardcoder-33b-v1.1.Q3_K_M.gguf	Q3_K_M	3	16.07 GB	18.57 GB	very small, high quality loss
wizardcoder-33b-v1.1.Q3_K_L.gguf	Q3_K_L	3	17.56 GB	20.06 GB	small, substantial quality loss
wizardcoder-33b-v1.1.Q4_0.gguf	Q4_0	4	18.82 GB	21.32 GB	legacy; small, very high quality loss - prefer using Q3_K_M
wizardcoder-33b-v1.1.Q4_K_S.gguf	Q4_K_S	4	18.89 GB	21.39 GB	small, greater quality loss
wizardcoder-33b-v1.1.Q4_K_M.gguf	Q4_K_M	4	19.94 GB	22.44 GB	medium, balanced quality - recommended
wizardcoder-33b-v1.1.Q5_0.gguf	Q5_0	5	22.96 GB	25.46 GB	legacy; medium, balanced quality - prefer using Q4_K_M
wizardcoder-33b-v1.1.Q5_K_S.gguf	Q5_K_S	5	22.96 GB	25.46 GB	large, low quality loss - recommended
wizardcoder-33b-v1.1.Q5_K_M.gguf	Q5_K_M	5	23.54 GB	26.04 GB	large, very low quality loss - recommended
wizardcoder-33b-v1.1.Q6_K.gguf	Q6_K	6	27.36 GB	29.86 GB	very large, extremely low quality loss
wizardcoder-33b-v1.1.Q8_0.gguf	Q8_0	8	35.43 GB	37.93 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/WizardCoder-33B-V1.1-GGUF and below it, a specific filename to download, such as: wizardcoder-33b-v1.1.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/WizardCoder-33B-V1.1-GGUF wizardcoder-33b-v1.1.Q4_K_M.gguf --local-dir . --local-dir-use-symlinks False

More advanced huggingface-cli download usage (click to read)

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

huggingface-cli download TheBloke/WizardCoder-33B-V1.1-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/WizardCoder-33B-V1.1-GGUF wizardcoder-33b-v1.1.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 35 -m wizardcoder-33b-v1.1.Q4_K_M.gguf --color -c 16384 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n\n### Instruction:\n{prompt}\n\n### Response:"

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

Change -c 16384 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. Note that longer sequence lengths require much more resources, so you may need to reduce this value.

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. Note that at the time of writing (Nov 27th 2023), ctransformers has not been updated for some time and is not compatible with some recent models. Therefore I recommend you use llama-cpp-python.

How to load this model in Python code, using llama-cpp-python

For full documentation, please see: llama-cpp-python docs.

First install the package

Run one of the following commands, according to your system:

shell

# Base ctransformers with no GPU acceleration
pip install llama-cpp-python
# With NVidia CUDA acceleration
CMAKE_ARGS="-DLLAMA_CUBLAS=on" pip install llama-cpp-python
# Or with OpenBLAS acceleration
CMAKE_ARGS="-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS" pip install llama-cpp-python
# Or with CLBLast acceleration
CMAKE_ARGS="-DLLAMA_CLBLAST=on" pip install llama-cpp-python
# Or with AMD ROCm GPU acceleration (Linux only)
CMAKE_ARGS="-DLLAMA_HIPBLAS=on" pip install llama-cpp-python
# Or with Metal GPU acceleration for macOS systems only
CMAKE_ARGS="-DLLAMA_METAL=on" pip install llama-cpp-python

# In windows, to set the variables CMAKE_ARGS in PowerShell, follow this format; eg for NVidia CUDA:
$env:CMAKE_ARGS = "-DLLAMA_OPENBLAS=on"
pip install llama-cpp-python

Simple llama-cpp-python example code

python

from llama_cpp import Llama

# 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 = Llama(
  model_path="./wizardcoder-33b-v1.1.Q4_K_M.gguf",  # Download the model file first
  n_ctx=16384,  # The max sequence length to use - note that longer sequence lengths require much more resources
  n_threads=8,            # The number of CPU threads to use, tailor to your system and the resulting performance
  n_gpu_layers=35         # The number of layers to offload to GPU, if you have GPU acceleration available
)

# Simple inference example
output = llm(
  "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n\n### Instruction:\n{prompt}\n\n### Response:", # Prompt
  max_tokens=512,  # Generate up to 512 tokens
  stop=["</s>"],   # Example stop token - not necessarily correct for this specific model! Please check before using.
  echo=True        # Whether to echo the prompt
)

# Chat Completion API

llm = Llama(model_path="./wizardcoder-33b-v1.1.Q4_K_M.gguf", chat_format="llama-2")  # Set chat_format according to the model you are using
llm.create_chat_completion(
    messages = [
        {"role": "system", "content": "You are a story writing assistant."},
        {
            "role": "user",
            "content": "Write a story about llamas."
        }
    ]
)

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:

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: Michael Levine, 阿明, Trailburnt, Nikolai Manek, John Detwiler, Randy H, Will Dee, Sebastain Graf, NimbleBox.ai, Eugene Pentland, Emad Mostaque, Ai Maven, Jim Angel, Jeff Scroggin, Michael Davis, Manuel Alberto Morcote, Stephen Murray, Robert, Justin Joy, Luke @flexchar, Brandon Frisco, Elijah Stavena, S_X, Dan Guido, Undi ., Komninos Chatzipapas, Shadi, theTransient, Lone Striker, Raven Klaugh, jjj, Cap'n Zoog, Michel-Marie MAUDET (LINAGORA), Matthew Berman, David, Fen Risland, Omer Bin Jawed, Luke Pendergrass, Kalila, OG, Erik Bjäreholt, Rooh Singh, Joseph William Delisle, Dan Lewis, TL, John Villwock, AzureBlack, Brad, Pedro Madruga, Caitlyn Gatomon, K, jinyuan sun, Mano Prime, Alex, Jeffrey Morgan, Alicia Loh, Illia Dulskyi, Chadd, transmissions 11, fincy, Rainer Wilmers, ReadyPlayerEmma, knownsqashed, Mandus, biorpg, Deo Leter, Brandon Phillips, SuperWojo, Sean Connelly, Iucharbius, Jack West, Harry Royden McLaughlin, Nicholas, terasurfer, Vitor Caleffi, Duane Dunston, Johann-Peter Hartmann, David Ziegler, Olakabola, Ken Nordquist, Trenton Dambrowitz, Tom X Nguyen, Vadim, Ajan Kanaga, Leonard Tan, Clay Pascal, Alexandros Triantafyllidis, JM33133, Xule, vamX, ya boyyy, subjectnull, Talal Aujan, Alps Aficionado, wassieverse, Ari Malik, James Bentley, Woland, Spencer Kim, Michael Dempsey, Fred von Graf, Elle, zynix, William Richards, Stanislav Ovsiannikov, Edmond Seymore, Jonathan Leane, Martin Kemka, usrbinkat, Enrico Ros

Thank you to all my generous patrons and donaters!

And thank you again to a16z for their generous grant.

Original model card: WizardLM's Wizardcoder 33B V1.1

WizardCoder: Empowering Code Large Language Models with Evol-Instruct

🏠 Home Page

🤗 HF Repo •🐱 Github Repo • 🐦 Twitter

📃 [WizardLM] • 📃 [WizardCoder] • 📃 [WizardMath]

👋 Join our Discord

News

[2023/01/04] 🔥 We released WizardCoder-33B-V1.1 trained from deepseek-coder-33b-base, the SOTA OSS Code LLM on EvalPlus Leaderboard, achieves 79.9 pass@1 on HumanEval, 73.2 pass@1 on HumanEval-Plus, 78.9 pass@1 on MBPP, and 66.9 pass@1 on MBPP-Plus.

[2023/01/04] 🔥 WizardCoder-33B-V1.1 outperforms ChatGPT 3.5, Gemini Pro, and DeepSeek-Coder-33B-instruct on HumanEval and HumanEval-Plus pass@1.

[2023/01/04] 🔥 WizardCoder-33B-V1.1 is comparable with ChatGPT 3.5, and surpasses Gemini Pro on MBPP and MBPP-Plus pass@1.

Model	Checkpoint	Paper	HumanEval	HumanEval+	MBPP	MBPP+	License
GPT-4-Turbo (Nov 2023)	-	-	85.4	81.7	83.0	70.7	-
GPT-4 (May 2023)	-	-	88.4	76.8	-	-	-
GPT-3.5-Turbo (Nov 2023)	-	-	72.6	65.9	81.7	69.4	-
Gemini Pro	-	-	63.4	55.5	72.9	57.9	-
DeepSeek-Coder-33B-instruct	-	-	78.7	72.6	78.7	66.7	-
WizardCoder-33B-V1.1	🤗 HF Link	📃 [WizardCoder]	79.9	73.2	78.9	66.9	MSFTResearch
WizardCoder-Python-34B-V1.0	🤗 HF Link	📃 [WizardCoder]	73.2	64.6	73.2	59.9	Llama2
WizardCoder-15B-V1.0	🤗 HF Link	📃 [WizardCoder]	59.8	52.4	--	--	OpenRAIL-M
WizardCoder-Python-13B-V1.0	🤗 HF Link	📃 [WizardCoder]	64.0	--	--	--	Llama2
WizardCoder-Python-7B-V1.0	🤗 HF Link	📃 [WizardCoder]	55.5	--	--	--	Llama2
WizardCoder-3B-V1.0	🤗 HF Link	📃 [WizardCoder]	34.8	--	--	--	OpenRAIL-M
WizardCoder-1B-V1.0	🤗 HF Link	📃 [WizardCoder]	23.8	--	--	--	OpenRAIL-M

❗ Data Contamination Check:

Before model training, we carefully and rigorously checked all the training data, and used multiple deduplication methods to verify and prevent data leakage on HumanEval and MBPP test set.

🔥 ❗Note for model system prompts usage:

Please use the same systems prompts strictly with us, and we do not guarantee the accuracy of the quantified versions.

Default version:

"Below is an instruction that describes a task. Write a response that appropriately completes the request.\n\n### Instruction:\n{instruction}\n\n### Response:"

How to Reproduce the Performance of WizardCoder-33B-V1.1

We provide all codes here.

We also provide all generated results.

transformers==4.36.2
vllm==0.2.5

(1) HumanEval and HumanEval-Plus

Step 1

Code Generation (w/o accelerate)

bash

model="WizardLM/WizardCoder-33B-V1.1"
temp=0.0
max_len=2048
pred_num=1
num_seqs_per_iter=1

output_path=preds/T${temp}_N${pred_num}_WizardCoder-33B-V1.1_Greedy_Decode

mkdir -p ${output_path}
echo 'Output path: '$output_path
echo 'Model to eval: '$model

# 164 problems, 21 per GPU if GPU=8
index=0
gpu_num=8
for ((i = 0; i < $gpu_num; i++)); do
  start_index=$((i * 21))
  end_index=$(((i + 1) * 21))

  gpu=$((i))
  echo 'Running process #' ${i} 'from' $start_index 'to' $end_index 'on GPU' ${gpu}
  ((index++))
  (
    CUDA_VISIBLE_DEVICES=$gpu python humaneval_gen.py --model ${model} \
      --start_index ${start_index} --end_index ${end_index} --temperature ${temp} \
      --num_seqs_per_iter ${num_seqs_per_iter} --N ${pred_num} --max_len ${max_len} --output_path ${output_path} --greedy_decode
  ) &
  if (($index % $gpu_num == 0)); then wait; fi
done

Code Generation (w/ vllm accelerate)

bash

model="WizardLM/WizardCoder-33B-V1.1"
temp=0.0
max_len=2048
pred_num=1
num_seqs_per_iter=1

output_path=preds/T${temp}_N${pred_num}_WizardCoder-33B-V1.1_Greedy_Decode_vllm

mkdir -p ${output_path}
echo 'Output path: '$output_path
echo 'Model to eval: '$model

CUDA_VISIBLE_DEVICES=0,1,2,3 python humaneval_gen_vllm.py --model ${model} \
    --start_index 0 --end_index 164 --temperature ${temp} \
    --num_seqs_per_iter ${num_seqs_per_iter} --N ${pred_num} --max_len ${max_len} --output_path ${output_path} --num_gpus 4 --overwrite

Step 2: Get the score

Install Eval-Plus benchmark.

bash

git clone https://github.com/evalplus/evalplus.git
cd evalplus
export PYTHONPATH=$PYTHONPATH:$(pwd)
pip install -r requirements.txt

Get HumanEval and HumanEval-Plus scores.

bash

output_path=preds/T0.0_N1_WizardCoder-33B-V1.1_Greedy_Decode

echo 'Output path: '$output_path
python process_humaneval.py --path ${output_path} --out_path ${output_path}.jsonl --add_prompt

evalplus.evaluate --dataset humaneval --samples ${output_path}.jsonl

(2) MBPP and MBPP-Plus

The preprocessed questions are provided in mbppplus.json.

Step 1

Code Generation (w/o accelerate)

bash

model="WizardLM/WizardCoder-33B-V1.1"
temp=0.0
max_len=2048
pred_num=1
num_seqs_per_iter=1

output_path=preds/MBPP_T${temp}_N${pred_num}_WizardCoder-33B-V1.1_Greedy_Decode

mkdir -p ${output_path}
echo 'Output path: '$output_path
echo 'Model to eval: '$model

# 399 problems, 50 per GPU if GPU=8
index=0
gpu_num=8
for ((i = 0; i < $gpu_num; i++)); do
  start_index=$((i * 50))
  end_index=$(((i + 1) * 50))

  gpu=$((i))
  echo 'Running process #' ${i} 'from' $start_index 'to' $end_index 'on GPU' ${gpu}
  ((index++))
  (
    CUDA_VISIBLE_DEVICES=$gpu python mbppplus_gen.py --model ${model} \
      --start_index ${start_index} --end_index ${end_index} --temperature ${temp} \
      --num_seqs_per_iter ${num_seqs_per_iter} --N ${pred_num} --max_len ${max_len} --output_path ${output_path} --mbpp_path "mbppplus.json" --greedy_decode
  ) &
  if (($index % $gpu_num == 0)); then wait; fi
done

Code Generation (w/ vllm accelerate)

bash

model="WizardLM/WizardCoder-33B-V1.1"
temp=0.0
max_len=2048
pred_num=1
num_seqs_per_iter=1

output_path=preds/MBPP_T${temp}_N${pred_num}_WizardCoder-33B-V1.1_Greedy_Decode_vllm

mkdir -p ${output_path}
echo 'Output path: '$output_path
echo 'Model to eval: '$model

CUDA_VISIBLE_DEVICES=0,1,2,3 python mbppplus_gen_vllm.py --model ${model} \
    --start_index ${start_index} --end_index ${end_index} --temperature ${temp} \
    --num_seqs_per_iter ${num_seqs_per_iter} --N ${pred_num} --max_len ${max_len} --output_path ${output_path} --mbpp_path "mbppplus.json" --num_gpus 4

Step 2: Get the score

Install Eval-Plus benchmark.

bash

git clone https://github.com/evalplus/evalplus.git
cd evalplus
export PYTHONPATH=$PYTHONPATH:$(pwd)
pip install -r requirements.txt

Get HumanEval and HumanEval-Plus scores.

bash

output_path=preds/MBPP_T0.0_N1_WizardCoder-33B-V1.1_Greedy_Decode

echo 'Output path: '$output_path
python mbppplus_process_preds.py --path ${output_path} --out_path ${output_path}.jsonl --add_prompt

evalplus.evaluate --dataset mbpp --samples ${output_path}.jsonl

Citation

Please cite the repo if you use the data, method or code in this repo.

@article{luo2023wizardcoder,
  title={WizardCoder: Empowering Code Large Language Models with Evol-Instruct},
  author={Luo, Ziyang and Xu, Can and Zhao, Pu and Sun, Qingfeng and Geng, Xiubo and Hu, Wenxiang and Tao, Chongyang and Ma, Jing and Lin, Qingwei and Jiang, Daxin},
  journal={arXiv preprint arXiv:2306.08568},
  year={2023}
}

TheBloke/WizardCoder-33B-V1.1-GGUF

Wizardcoder 33B V1.1 - GGUF

Description

About GGUF

Repositories available

Prompt template: Alpaca

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 in Python code, using llama-cpp-python

First install the package

Simple llama-cpp-python example code

How to use with LangChain

Discord

Thanks, and how to contribute

Original model card: WizardLM's Wizardcoder 33B V1.1

WizardCoder: Empowering Code Large Language Models with Evol-Instruct

News

❗ Data Contamination Check:

How to Reproduce the Performance of WizardCoder-33B-V1.1

Citation

TheBloke/
WizardCoder-33B-V1.1-GGUF

In `text-generation-webui`

Example `llama.cpp` command

How to run in `text-generation-webui`