* Large number of extensions (built-in and user-contributed), including Coqui TTS for realistic voice outputs, Whisper STT for voice inputs, translation, [multimodal pipelines](https://github.com/oobabooga/text-generation-webui/tree/main/extensions/multimodal), vector databases, Stable Diffusion integration, and a lot more. See [the wiki](https://github.com/oobabooga/text-generation-webui/wiki/07-%E2%80%90-Extensions) and [the extensions directory](https://github.com/oobabooga/text-generation-webui-extensions) for details.
* Precise templates for instruction-following models, including Llama-2-chat, Alpaca, Vicuna, Mistral, and many others.
* Easy UI for training LoRAs, as well as loading/unloading them on the fly.
* HF transformers integration: load models in 4-bit or 8-bit quantization 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](https://github.com/oobabooga/text-generation-webui/wiki/12-%E2%80%90-OpenAI-API#examples).
To restart the web UI later, just run the `start_` script once again. This script will set up the necessary files for the web UI in the `installer_files` folder. In case you need to reinstall the project's requirements, you can easily do so by deleting that folder and then running the `start_` script again.
You also have the option of using command-line flags with the script. As an alternative, you can edit the `CMD_FLAGS.txt` file with a text editor and include your flags there.
If you ever need to install something manually in the `installer_files` environment, you can launch an interactive shell using the cmd script: `cmd_linux.sh`, `cmd_windows.bat`, `cmd_macos.sh`, or `cmd_wsl.bat`.
* The installer has been tested mostly on NVIDIA GPUs. If you can find a way to improve it for your AMD/Intel Arc/Mac Metal GPU, you are highly encouraged to submit a PR to this repository. The main file to be edited is `one_click.py`.
* 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 ./start_linux.sh`.
2) Manually install llama-cpp-python using the appropriate command for your hardware: [Installation from PyPI](https://github.com/abetlen/llama-cpp-python#installation-with-hardware-acceleration).
* Use the `LLAMA_HIPBLAS=on` toggle.
* Note the [Windows remarks](https://github.com/abetlen/llama-cpp-python#windows-remarks).
4) Manually install [ExLlama](https://github.com/turboderp/exllama) by simply cloning it into the `repositories` folder (it will be automatically compiled at runtime after that):
The requirements*.txt above contain various precompiled wheels. 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.
* You need to have Docker Compose v2.17 or higher installed. See [this guide](https://github.com/oobabooga/text-generation-webui/wiki/09-%E2%80%90-Docker) for instructions.
| `-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. |
| `--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. |
#### Accelerate 4-bit
⚠️ Requires minimum compute of 7.0 on Windows at the moment.
| `--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. |
| `--cache-capacity CACHE_CAPACITY` | Maximum cache capacity (llama-cpp-python). Examples: 2000MiB, 2GiB. When provided without units, bytes will be assumed. |
#### ExLlama
| 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. |
#### AutoGPTQ
| 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. |
#### GPTQ-for-LLaMa
| 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. |
#### ctransformers
| 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. |
| `--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`. |
| `--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. |
Models should be placed in the folder `text-generation-webui/models`. They are usually downloaded from [Hugging Face](https://huggingface.co/models?pipeline_tag=text-generation&sort=downloads).
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 via the command-line with
If you would like to contribute to the project, check out the [Contributing guidelines](https://github.com/oobabooga/text-generation-webui/wiki/Contributing-guidelines).
In August 2023, [Andreessen Horowitz](https://a16z.com/) (a16z) provided a generous grant to encourage and support my independent work on this project. I am **extremely** grateful for their trust and recognition, which will allow me to dedicate more time towards realizing the full potential of text-generation-webui.
If you find this project useful, I have a [Ko-fi page](https://ko-fi.com/oobabooga) where you can make a donation. Your support helps me continue maintaining and improving this project.
