Skip to content

Commit 3da73e4

Browse files
oobaboogaoobabooga
committed
Merge branch 'main' into Zerogoki00-opt4-bit
2 parents 518e5c4 + b327554 commit 3da73e4

File tree

11 files changed

+159
-50
lines changed

11 files changed

+159
-50
lines changed

‎.github/FUNDING.yml

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1 @@
1+
ko_fi: oobabooga

‎.github/ISSUE_TEMPLATE/bug_report_template.yml

Lines changed: 53 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,53 @@
1+
name: "Bug report"
2+
description: Report a bug
3+
labels: [ "bug" ]
4+
body:
5+
- type: markdown
6+
attributes:
7+
value: |
8+
Thanks for taking the time to fill out this bug report!
9+
- type: textarea
10+
id: bug-description
11+
attributes:
12+
label: Describe the bug
13+
description: A clear and concise description of what the bug is.
14+
placeholder: Bug description
15+
validations:
16+
required: true
17+
- type: checkboxes
18+
attributes:
19+
label: Is there an existing issue for this?
20+
description: Please search to see if an issue already exists for the issue you encountered.
21+
options:
22+
- label: I have searched the existing issues
23+
required: true
24+
- type: textarea
25+
id: reproduction
26+
attributes:
27+
label: Reproduction
28+
description: Please provide the steps necessary to reproduce your issue.
29+
placeholder: Reproduction
30+
validations:
31+
required: true
32+
- type: textarea
33+
id: screenshot
34+
attributes:
35+
label: Screenshot
36+
description: "If possible, please include screenshot(s) so that we can understand what the issue is."
37+
- type: textarea
38+
id: logs
39+
attributes:
40+
label: Logs
41+
description: "Please include the full stacktrace of the errors you get in the command-line (if any)."
42+
render: shell
43+
validations:
44+
required: true
45+
- type: textarea
46+
id: system-info
47+
attributes:
48+
label: System Info
49+
description: "Please share your system info with us: operating system, GPU brand, and GPU model. If you are using a Google Colab notebook, mention that instead."
50+
render: shell
51+
placeholder:
52+
validations:
53+
required: true

‎.github/ISSUE_TEMPLATE/feature_request.md

Lines changed: 16 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,16 @@
1+
---
2+
name: Feature request
3+
about: Suggest an improvement or new feature for the web UI
4+
title: ''
5+
labels: 'enhancement'
6+
assignees: ''
7+
8+
---
9+
10+
**Description**
11+
12+
A clear and concise description of what you want to be implemented.
13+
14+
**Additional Context**
15+
16+
If applicable, please provide any extra information, external links, or screenshots that could be useful.

‎.github/dependabot.yml

Lines changed: 11 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,11 @@
1+
# To get started with Dependabot version updates, you'll need to specify which
2+
# package ecosystems to update and where the package manifests are located.
3+
# Please see the documentation for all configuration options:
4+
# https://docs.github.com/github/administering-a-repository/configuration-options-for-dependency-updates
5+
6+
version: 2
7+
updates:
8+
- package-ecosystem: "pip" # See documentation for possible values
9+
directory: "/" # Location of package manifests
10+
schedule:
11+
interval: "weekly"

‎.github/workflows/stale.yml

Lines changed: 22 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,22 @@
1+
name: Close inactive issues
2+
on:
3+
schedule:
4+
- cron: "10 23 * * *"
5+
6+
jobs:
7+
close-issues:
8+
runs-on: ubuntu-latest
9+
permissions:
10+
issues: write
11+
pull-requests: write
12+
steps:
13+
- uses: actions/stale@v5
14+
with:
15+
stale-issue-message: ""
16+
close-issue-message: "This issue has been closed due to inactivity for 30 days. If you believe it is still relevant, you can reopen it (if you are the author) or leave a comment below."
17+
days-before-issue-stale: 30
18+
days-before-issue-close: 0
19+
stale-issue-label: "stale"
20+
days-before-pr-stale: -1
21+
days-before-pr-close: -1
22+
repo-token: ${{ secrets.GITHUB_TOKEN }}

‎README.md

Lines changed: 36 additions & 35 deletions
Original file line numberDiff line numberDiff line change
@@ -60,8 +60,7 @@ pip3 install torch torchvision torchaudio --extra-index-url https://download.pyt
6060
conda install pytorch torchvision torchaudio git -c pytorch
6161
```
6262

63-
See also: [Installation instructions for human beings
64-
](https://github.com/oobabooga/text-generation-webui/wiki/Installation-instructions-for-human-beings)
63+
See also: [Installation instructions for human beings](https://github.com/oobabooga/text-generation-webui/wiki/Installation-instructions-for-human-beings).
6564

6665
## Installation option 2: one-click installers
6766

@@ -132,39 +131,41 @@ Then browse to
132131

133132
Optionally, you can use the following command-line flags:
134133

135-
| Flag | Description |
136-
|--------------------------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
137-
| `-h`, `--help` | show this help message and exit |
138-
| `--model MODEL` | Name of the model to load by default. |
139-
| `--notebook` | Launch the web UI in notebook mode, where the output is written to the same text box as the input. |
140-
| `--chat` | Launch the web UI in chat mode. |
141-
| `--cai-chat` | Launch the web UI in chat mode with a style similar to Character.AI's. If the file `img_bot.png` or `img_bot.jpg` exists in the same folder as server.py, this image will be used as the bot's profile picture. Similarly, `img_me.png` or `img_me.jpg` will be used as your profile picture. |
142-
| `--cpu` | Use the CPU to generate text. |
143-
| `--load-in-8bit` | Load the model with 8-bit precision. |
144-
| `--gptq-bits GPTQ_BITS` | Load a pre-quantized model with specified precision. 2, 3, 4 and 8 (bit) are supported. Currently only works with LLaMA and OPT. |
145-
| `--gptq-model-type MODEL_TYPE` | Model type of pre-quantized model. Currently only LLaMa and OPT are supported. |
146-
| `--bf16` | Load the model with bfloat16 precision. Requires NVIDIA Ampere GPU. |
147-
| `--auto-devices` | Automatically split the model across the available GPU(s) and CPU. |
148-
| `--disk` | If the model is too large for your GPU(s) and CPU combined, send the remaining layers to the disk. |
149-
| `--disk-cache-dir DISK_CACHE_DIR` | Directory to save the disk cache to. Defaults to `cache/`. |
150-
| `--gpu-memory GPU_MEMORY [GPU_MEMORY ...]` | Maxmimum GPU memory in GiB to be allocated per GPU. Example: `--gpu-memory 10` for a single GPU, `--gpu-memory 10 5` for two GPUs. |
151-
| `--cpu-memory CPU_MEMORY` | Maximum CPU memory in GiB to allocate for offloaded weights. Must be an integer number. Defaults to 99. |
152-
| `--flexgen` | Enable the use of FlexGen offloading. |
153-
| `--percent PERCENT [PERCENT ...]` | FlexGen: allocation percentages. Must be 6 numbers separated by spaces (default: 0, 100, 100, 0, 100, 0). |
154-
| `--compress-weight` | FlexGen: Whether to compress weight (default: False). |
155-
| `--pin-weight [PIN_WEIGHT]` | FlexGen: whether to pin weights (setting this to False reduces CPU memory by 20%). |
156-
| `--deepspeed` | Enable the use of DeepSpeed ZeRO-3 for inference via the Transformers integration. |
157-
| `--nvme-offload-dir NVME_OFFLOAD_DIR` | DeepSpeed: Directory to use for ZeRO-3 NVME offloading. |
158-
| `--local_rank LOCAL_RANK` | DeepSpeed: Optional argument for distributed setups. |
159-
| `--rwkv-strategy RWKV_STRATEGY` | RWKV: The strategy to use while loading the model. Examples: "cpu fp32", "cuda fp16", "cuda fp16i8". |
160-
| `--rwkv-cuda-on` | RWKV: Compile the CUDA kernel for better performance. |
161-
| `--no-stream` | Don't stream the text output in real time. This improves the text generation performance. |
162-
| `--settings SETTINGS_FILE` | Load the default interface settings from this json file. See `settings-template.json` for an example. If you create a file called `settings.json`, this file will be loaded by default without the need to use the `--settings` flag. |
163-
| `--extensions EXTENSIONS [EXTENSIONS ...]` | The list of extensions to load. If you want to load more than one extension, write the names separated by spaces. |
164-
| `--listen` | Make the web UI reachable from your local network. |
165-
| `--listen-port LISTEN_PORT` | The listening port that the server will use. |
166-
| `--share` | Create a public URL. This is useful for running the web UI on Google Colab or similar. |
167-
| `--verbose` | Print the prompts to the terminal. |
134+
| Flag | Description |
135+
|-------------|-------------|
136+
| `-h`, `--help` | show this help message and exit |
137+
| `--model MODEL` | Name of the model to load by default. |
138+
| `--notebook` | Launch the web UI in notebook mode, where the output is written to the same text box as the input. |
139+
| `--chat` | Launch the web UI in chat mode.|
140+
| `--cai-chat` | Launch the web UI in chat mode with a style similar to Character.AI's. If the file `img_bot.png` or `img_bot.jpg` exists in the same folder as server.py, this image will be used as the bot's profile picture. Similarly, `img_me.png` or `img_me.jpg` will be used as your profile picture. |
141+
| `--cpu` | Use the CPU to generate text.|
142+
| `--load-in-8bit` | Load the model with 8-bit precision.|
143+
| `--load-in-4bit` | Load the model with 4-bit precision. Currently only works with LLaMA.|
144+
| `--gptq-bits GPTQ_BITS` | Load a pre-quantized model with specified precision. 2, 3, 4 and 8 (bit) are supported. Currently only works with LLaMA and OPT. |
145+
| `--gptq-model-type MODEL_TYPE` | Model type of pre-quantized model. Currently only LLaMa and OPT are supported. |
146+
| `--bf16` | Load the model with bfloat16 precision. Requires NVIDIA Ampere GPU. |
147+
| `--auto-devices` | Automatically split the model across the available GPU(s) and CPU.|
148+
| `--disk` | If the model is too large for your GPU(s) and CPU combined, send the remaining layers to the disk. |
149+
| `--disk-cache-dir DISK_CACHE_DIR` | Directory to save the disk cache to. Defaults to `cache/`. |
150+
| `--gpu-memory GPU_MEMORY [GPU_MEMORY ...]` | Maxmimum GPU memory in GiB to be allocated per GPU. Example: `--gpu-memory 10` for a single GPU, `--gpu-memory 10 5` for two GPUs. |
151+
| `--cpu-memory CPU_MEMORY` | Maximum CPU memory in GiB to allocate for offloaded weights. Must be an integer number. Defaults to 99.|
152+
| `--flexgen` | Enable the use of FlexGen offloading. |
153+
| `--percent PERCENT [PERCENT ...]` | FlexGen: allocation percentages. Must be 6 numbers separated by spaces (default: 0, 100, 100, 0, 100, 0). |
154+
| `--compress-weight` | FlexGen: Whether to compress weight (default: False).|
155+
| `--pin-weight [PIN_WEIGHT]` | FlexGen: whether to pin weights (setting this to False reduces CPU memory by 20%). |
156+
| `--deepspeed` | Enable the use of DeepSpeed ZeRO-3 for inference via the Transformers integration. |
157+
| `--nvme-offload-dir NVME_OFFLOAD_DIR` | DeepSpeed: Directory to use for ZeRO-3 NVME offloading. |
158+
| `--local_rank LOCAL_RANK` | DeepSpeed: Optional argument for distributed setups. |
159+
| `--rwkv-strategy RWKV_STRATEGY` | RWKV: The strategy to use while loading the model. Examples: "cpu fp32", "cuda fp16", "cuda fp16i8". |
160+
| `--rwkv-cuda-on` | RWKV: Compile the CUDA kernel for better performance. |
161+
| `--no-stream` | Don't stream the text output in real time. |
162+
| `--settings SETTINGS_FILE` | Load the default interface settings from this json file. See `settings-template.json` for an example. If you create a file called `settings.json`, this file will be loaded by default without the need to use the `--settings` flag.|
163+
| `--extensions EXTENSIONS [EXTENSIONS ...]` | The list of extensions to load. If you want to load more than one extension, write the names separated by spaces. |
164+
| `--listen` | Make the web UI reachable from your local network.|
165+
| `--listen-port LISTEN_PORT` | The listening port that the server will use. |
166+
| `--share` | Create a public URL. This is useful for running the web UI on Google Colab or similar. |
167+
| `--auto-launch` | Open the web UI in the default browser upon launch. |
168+
| `--verbose` | Print the prompts to the terminal. |
168169

169170
Out of memory errors? [Check this guide](https://github.com/oobabooga/text-generation-webui/wiki/Low-VRAM-guide).
170171

0 commit comments

Comments
 (0)