StefanDanielSchwarz / text-generation-webui

A gradio web UI for running Large Language Models like LLaMA, llama.cpp, GPT-J, OPT, and GALACTICA.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Text generation web UI

A Gradio web UI for Large Language Models.

Its goal is to become the AUTOMATIC1111/stable-diffusion-webui of text generation.

Image1 Image2
Image1 Image2


  • 3 interface modes: default (two columns), notebook, and chat.
  • Multiple model backends: Transformers, llama.cpp (through llama-cpp-python), ExLlama, ExLlamaV2, AutoGPTQ, AutoAWQ, GPTQ-for-LLaMa, CTransformers, QuIP#.
  • Dropdown menu for quickly switching between different models.
  • Large number of extensions (built-in and user-contributed), including Coqui TTS for realistic voice outputs, Whisper STT for voice inputs, translation, multimodal pipelines, vector databases, Stable Diffusion integration, and a lot more. See the wiki and the extensions directory for details.
  • Chat with custom characters.
  • Precise chat templates for instruction-following models, including Llama-2-chat, Alpaca, Vicuna, Mistral.
  • LoRA: train new LoRAs with your own data, load/unload LoRAs on the fly for generation.
  • Transformers library integration: load models in 4-bit or 8-bit precision through bitsandbytes, use llama.cpp with transformers samplers (llamacpp_HF loader), CPU inference in 32-bit precision using PyTorch.
  • OpenAI-compatible API server with Chat and Completions endpoints -- see the examples.

How to install

  1. Clone or download the repository.
  2. Run the, start_windows.bat,, or start_wsl.bat script depending on your OS.
  3. Select your GPU vendor when asked.
  4. Once the installation ends, browse to http://localhost:7860/?__theme=dark.
  5. Have fun!

To restart the web UI in the future, just run the start_ script again. This script creates an installer_files folder where it sets up the project's requirements. In case you need to reinstall the requirements, you can simply delete that folder and start the web UI again.

The script accepts command-line flags. Alternatively, you can edit the CMD_FLAGS.txt file with a text editor and add your flags there.

To get updates in the future, run, update_windows.bat,, or update_wsl.bat.

Setup details and information about installing manually


The script uses Miniconda to set up a Conda environment in the installer_files folder.

If you ever need to install something manually in the installer_files environment, you can launch an interactive shell using the cmd script:, cmd_windows.bat,, or cmd_wsl.bat.

  • There is no need to run any of those scripts (start_, update_, or cmd_) as admin/root.
  • For additional instructions about AMD and WSL setup, consult the documentation.
  • For automated installation, you can use the GPU_CHOICE, USE_CUDA118, LAUNCH_AFTER_INSTALL, and INSTALL_EXTENSIONS environment variables. For instance: GPU_CHOICE=A USE_CUDA118=FALSE LAUNCH_AFTER_INSTALL=FALSE INSTALL_EXTENSIONS=FALSE ./

Manual installation using Conda

Recommended if you have some experience with the command-line.

0. Install Conda

On Linux or WSL, it can be automatically installed with these two commands (source):

curl -sL "" > ""

1. Create a new conda environment

conda create -n textgen python=3.11
conda activate textgen

2. Install Pytorch

System GPU Command
Linux/WSL NVIDIA pip3 install torch torchvision torchaudio --index-url
Linux/WSL CPU only pip3 install torch torchvision torchaudio --index-url
Linux AMD pip3 install torch torchvision torchaudio --index-url
MacOS + MPS Any pip3 install torch torchvision torchaudio
Windows NVIDIA pip3 install torch torchvision torchaudio --index-url
Windows CPU only pip3 install torch torchvision torchaudio

The up-to-date commands can be found here:

For NVIDIA, you also need to install the CUDA runtime libraries:

conda install -y -c "nvidia/label/cuda-12.1.1" cuda-runtime

If you need nvcc to compile some library manually, replace the command above with

conda install -y -c "nvidia/label/cuda-12.1.1" cuda

3. Install the web UI

git clone
cd text-generation-webui
pip install -r <requirements file according to table below>

Requirements file to use:

GPU CPU requirements file to use
NVIDIA has AVX2 requirements.txt
NVIDIA no AVX2 requirements_noavx2.txt
AMD has AVX2 requirements_amd.txt
AMD no AVX2 requirements_amd_noavx2.txt
CPU only has AVX2 requirements_cpu_only.txt
CPU only no AVX2 requirements_cpu_only_noavx2.txt
Apple Intel requirements_apple_intel.txt
Apple Apple Silicon requirements_apple_silicon.txt

Start the web UI

conda activate textgen
cd text-generation-webui

Then browse to


AMD GPU on Windows
  1. Use requirements_cpu_only.txt or requirements_cpu_only_noavx2.txt in the command above.

  2. Manually install llama-cpp-python using the appropriate command for your hardware: Installation from PyPI.

  3. Manually install AutoGPTQ: Installation.

    • Perform the from-source installation - there are no prebuilt ROCm packages for Windows.
  4. Manually install ExLlama by simply cloning it into the repositories folder (it will be automatically compiled at runtime after that):

cd text-generation-webui
git clone repositories/exllama
  1. For Kepler GPUs and older, you will need to install CUDA 11.8 instead of 12:
pip3 install torch torchvision torchaudio --index-url
conda install -y -c "nvidia/label/cuda-11.8.0" cuda-runtime
  1. bitsandbytes >= 0.39 may not work. In that case, to use --load-in-8bit, you may have to downgrade like this:
    • Linux: pip install bitsandbytes==0.38.1
    • Windows: pip install
Manual install

The requirements*.txt above contain various wheels precompiled through GitHub Actions. If you wish to compile things manually, or if you need to because no suitable wheels are available for your hardware, you can use requirements_nowheels.txt and then install your desired loaders manually.

Alternative: Docker

ln -s docker/{nvidia/Dockerfile,docker-compose.yml,.dockerignore} .
cp docker/.env.example .env
# Edit .env and set: 
#   TORCH_CUDA_ARCH_LIST based on your GPU model
#   APP_RUNTIME_GID      your host user's group id (run `id -g` in a terminal)
#   BUILD_EXTENIONS      optionally add comma separated list of extensions to build
docker compose up --build
  • You need to have Docker Compose v2.17 or higher installed. See this guide for instructions.
  • For additional docker files, check out this repository.

Updating the requirements

From time to time, the requirements*.txt change. To update, use these commands:

conda activate textgen
cd text-generation-webui
pip install -r <requirements file that you have used> --upgrade
List of command-line flags

Basic settings

Flag Description
-h, --help show this help message and exit
--multi-user Multi-user mode. Chat histories are not saved or automatically loaded. WARNING: this is likely not safe for sharing publicly.
--character CHARACTER The name of the character to load in chat mode by default.
--model MODEL Name of the model to load by default.
--lora LORA [LORA ...] The list of LoRAs to load. If you want to load more than one LoRA, write the names separated by spaces.
--model-dir MODEL_DIR Path to directory with all the models.
--lora-dir LORA_DIR Path to directory with all the loras.
--model-menu Show a model menu in the terminal when the web UI is first launched.
--settings SETTINGS_FILE Load the default interface settings from this yaml file. See settings-template.yaml for an example. If you create a file called settings.yaml, this file will be loaded by default without the need to use the --settings flag.
--extensions EXTENSIONS [EXTENSIONS ...] The list of extensions to load. If you want to load more than one extension, write the names separated by spaces.
--verbose Print the prompts to the terminal.
--chat-buttons Show buttons on the chat tab instead of a hover menu.

Model loader

Flag Description
--loader LOADER Choose the model loader manually, otherwise, it will get autodetected. Valid options: Transformers, llama.cpp, llamacpp_HF, ExLlama_HF, ExLlamav2_HF, AutoGPTQ, AutoAWQ, GPTQ-for-LLaMa, ExLlama, ExLlamav2, ctransformers, QuIP#.


Flag Description
--cpu Use the CPU to generate text. Warning: Training on CPU is extremely slow.
--auto-devices Automatically split the model across the available GPU(s) and CPU.
--gpu-memory GPU_MEMORY [GPU_MEMORY ...] Maximum GPU memory in GiB to be allocated per GPU. Example: --gpu-memory 10 for a single GPU, --gpu-memory 10 5 for two GPUs. You can also set values in MiB like --gpu-memory 3500MiB.
--cpu-memory CPU_MEMORY Maximum CPU memory in GiB to allocate for offloaded weights. Same as above.
--disk If the model is too large for your GPU(s) and CPU combined, send the remaining layers to the disk.
--disk-cache-dir DISK_CACHE_DIR Directory to save the disk cache to. Defaults to "cache".
--load-in-8bit Load the model with 8-bit precision (using bitsandbytes).
--bf16 Load the model with bfloat16 precision. Requires NVIDIA Ampere GPU.
--no-cache Set use_cache to False while generating text. This reduces VRAM usage slightly, but it comes at a performance cost.
--xformers Use xformer's memory efficient attention. This is really old and probably doesn't do anything.
--sdp-attention Use PyTorch 2.0's SDP attention. Same as above.
--trust-remote-code Set trust_remote_code=True while loading the model. Necessary for some models.
--no_use_fast Set use_fast=False while loading the tokenizer (it's True by default). Use this if you have any problems related to use_fast.
--use_flash_attention_2 Set use_flash_attention_2=True while loading the model.

bitsandbytes 4-bit

⚠️ Requires minimum compute of 7.0 on Windows at the moment.

Flag Description
--load-in-4bit Load the model with 4-bit precision (using bitsandbytes).
--use_double_quant use_double_quant for 4-bit.
--compute_dtype COMPUTE_DTYPE compute dtype for 4-bit. Valid options: bfloat16, float16, float32.
--quant_type QUANT_TYPE quant_type for 4-bit. Valid options: nf4, fp4.


Flag Description
--tensorcores Use llama-cpp-python compiled with tensor cores support. This increases performance on RTX cards. NVIDIA only.
--n_ctx N_CTX Size of the prompt context.
--threads Number of threads to use.
--threads-batch THREADS_BATCH Number of threads to use for batches/prompt processing.
--no_mul_mat_q Disable the mulmat kernels.
--n_batch Maximum number of prompt tokens to batch together when calling llama_eval.
--no-mmap Prevent mmap from being used.
--mlock Force the system to keep the model in RAM.
--n-gpu-layers N_GPU_LAYERS Number of layers to offload to the GPU.
--tensor_split TENSOR_SPLIT Split the model across multiple GPUs. Comma-separated list of proportions. Example: 18,17.
--numa Activate NUMA task allocation for llama.cpp.
--logits_all Needs to be set for perplexity evaluation to work. Otherwise, ignore it, as it makes prompt processing slower.
--no_offload_kqv Do not offload the K, Q, V to the GPU. This saves VRAM but reduces the performance.
--cache-capacity CACHE_CAPACITY Maximum cache capacity (llama-cpp-python). Examples: 2000MiB, 2GiB. When provided without units, bytes will be assumed.


Flag Description
--gpu-split Comma-separated list of VRAM (in GB) to use per GPU device for model layers. Example: 20,7,7.
--max_seq_len MAX_SEQ_LEN Maximum sequence length.
--cfg-cache ExLlama_HF: Create an additional cache for CFG negative prompts. Necessary to use CFG with that loader, but not necessary for CFG with base ExLlama.
--no_flash_attn Force flash-attention to not be used.
--cache_8bit Use 8-bit cache to save VRAM.
--num_experts_per_token NUM_EXPERTS_PER_TOKEN Number of experts to use for generation. Applies to MoE models like Mixtral.


Flag Description
--triton Use triton.
--no_inject_fused_attention Disable the use of fused attention, which will use less VRAM at the cost of slower inference.
--no_inject_fused_mlp Triton mode only: disable the use of fused MLP, which will use less VRAM at the cost of slower inference.
--no_use_cuda_fp16 This can make models faster on some systems.
--desc_act For models that don't have a quantize_config.json, this parameter is used to define whether to set desc_act or not in BaseQuantizeConfig.
--disable_exllama Disable ExLlama kernel, which can improve inference speed on some systems.
--disable_exllamav2 Disable ExLlamav2 kernel.


Flag Description
--wbits WBITS Load a pre-quantized model with specified precision in bits. 2, 3, 4 and 8 are supported.
--model_type MODEL_TYPE Model type of pre-quantized model. Currently LLaMA, OPT, and GPT-J are supported.
--groupsize GROUPSIZE Group size.
--pre_layer PRE_LAYER [PRE_LAYER ...] The number of layers to allocate to the GPU. Setting this parameter enables CPU offloading for 4-bit models. For multi-gpu, write the numbers separated by spaces, eg --pre_layer 30 60.
--checkpoint CHECKPOINT The path to the quantized checkpoint file. If not specified, it will be automatically detected.
--monkey-patch Apply the monkey patch for using LoRAs with quantized models.


Flag Description
--model_type MODEL_TYPE Model type of pre-quantized model. Currently gpt2, gptj, gptneox, falcon, llama, mpt, starcoder (gptbigcode), dollyv2, and replit are supported.


Flag Description
--hqq-backend Backend for the HQQ loader. Valid options: PYTORCH, PYTORCH_COMPILE, ATEN.


Flag Description
--deepspeed Enable the use of DeepSpeed ZeRO-3 for inference via the Transformers integration.
--nvme-offload-dir NVME_OFFLOAD_DIR DeepSpeed: Directory to use for ZeRO-3 NVME offloading.
--local_rank LOCAL_RANK DeepSpeed: Optional argument for distributed setups.


Flag Description
--rwkv-strategy RWKV_STRATEGY RWKV: The strategy to use while loading the model. Examples: "cpu fp32", "cuda fp16", "cuda fp16i8".
--rwkv-cuda-on RWKV: Compile the CUDA kernel for better performance.

RoPE (for llama.cpp, ExLlama, ExLlamaV2, and transformers)

Flag Description
--alpha_value ALPHA_VALUE Positional embeddings alpha factor for NTK RoPE scaling. Use either this or compress_pos_emb, not both.
--rope_freq_base ROPE_FREQ_BASE If greater than 0, will be used instead of alpha_value. Those two are related by rope_freq_base = 10000 * alpha_value ^ (64 / 63).
--compress_pos_emb COMPRESS_POS_EMB Positional embeddings compression factor. Should be set to (context length) / (model's original context length). Equal to 1/rope_freq_scale.


Flag Description
--listen Make the web UI reachable from your local network.
--listen-port LISTEN_PORT The listening port that the server will use.
--listen-host LISTEN_HOST The hostname that the server will use.
--share Create a public URL. This is useful for running the web UI on Google Colab or similar.
--auto-launch Open the web UI in the default browser upon launch.
--gradio-auth USER:PWD Set Gradio authentication password in the format "username:password". Multiple credentials can also be supplied with "u1:p1,u2:p2,u3:p3".
--gradio-auth-path GRADIO_AUTH_PATH Set the Gradio authentication file path. The file should contain one or more user:password pairs in the same format as above.
--ssl-keyfile SSL_KEYFILE The path to the SSL certificate key file.
--ssl-certfile SSL_CERTFILE The path to the SSL certificate cert file.


Flag Description
--api Enable the API extension.
--public-api Create a public URL for the API using Cloudfare.
--public-api-id PUBLIC_API_ID Tunnel ID for named Cloudflare Tunnel. Use together with public-api option.
--api-port API_PORT The listening port for the API.
--api-key API_KEY API authentication key.
--admin-key ADMIN_KEY API authentication key for admin tasks like loading and unloading models. If not set, will be the same as --api-key.
--nowebui Do not launch the Gradio UI. Useful for launching the API in standalone mode.


Flag Description
--multimodal-pipeline PIPELINE The multimodal pipeline to use. Examples: llava-7b, llava-13b.


Downloading models

Models should be placed in the folder text-generation-webui/models. They are usually downloaded from Hugging Face.

  • GGUF models are a single file and should be placed directly into models. Example:
└── models
    └── llama-2-13b-chat.Q4_K_M.gguf
  • The remaining model types (like 16-bit transformers models and GPTQ models) are made of several files and must be placed in a subfolder. Example:
├── models
│   ├── lmsys_vicuna-33b-v1.3
│   │   ├── config.json
│   │   ├── generation_config.json
│   │   ├── pytorch_model-00001-of-00007.bin
│   │   ├── pytorch_model-00002-of-00007.bin
│   │   ├── pytorch_model-00003-of-00007.bin
│   │   ├── pytorch_model-00004-of-00007.bin
│   │   ├── pytorch_model-00005-of-00007.bin
│   │   ├── pytorch_model-00006-of-00007.bin
│   │   ├── pytorch_model-00007-of-00007.bin
│   │   ├── pytorch_model.bin.index.json
│   │   ├── special_tokens_map.json
│   │   ├── tokenizer_config.json
│   │   └── tokenizer.model

In both cases, you can use the "Model" tab of the UI to download the model from Hugging Face automatically. It is also possible to download it via the command-line with

python organization/model

Run python --help to see all the options.

Google Colab notebook


If you would like to contribute to the project, check out the Contributing guidelines.


Acknowledgment & support

In August 2023, Andreessen Horowitz (a16z) provided a generous grant to encourage and support my independent work on this project. I am extremely grateful for their trust and recognition.

If you find this project useful, I have a Ko-fi page where you can make a donation. Your support helps me continue maintaining and improving this project.


A gradio web UI for running Large Language Models like LLaMA, llama.cpp, GPT-J, OPT, and GALACTICA.

License:GNU Affero General Public License v3.0


Language:Python 91.9%Language:CSS 3.4%Language:JavaScript 1.9%Language:Shell 1.3%Language:Batchfile 0.7%Language:Jupyter Notebook 0.6%Language:Dockerfile 0.3%