Skip to content

Commit

Permalink
Merge pull request #11 from Infotrend-Inc/o1-wip
Browse files Browse the repository at this point in the history
0.9.10
  • Loading branch information
mmartial authored Dec 17, 2024
2 parents bed5223 + 2fc3c68 commit 44c721a
Show file tree
Hide file tree
Showing 6 changed files with 37 additions and 9 deletions.
2 changes: 1 addition & 1 deletion Makefile
Original file line number Diff line number Diff line change
Expand Up @@ -15,7 +15,7 @@ DOCKER_CMD=docker
OAI_BASE="python:3.12-slim-bookworm"

# Version infomation and container name
OAI_VERSION="0.9.9"
OAI_VERSION="0.9.10"

OAI_CONTAINER_NAME="openai_webui"

Expand Down
10 changes: 8 additions & 2 deletions README.md
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
<h1>OpenAI WebUI</h1>

Latest version: 0.9.9 (20241206)
Latest version: 0.9.10 (20241217)

- [1. Description](#1-description)
- [1.1. Supported models](#11-supported-models)
Expand Down Expand Up @@ -104,9 +104,14 @@ The following table shows the [models](https://platform.openai.com/docs/models/)
| GPT | o1-preview-2024-09-12 | active | | `beta` | 0.9.9 |
| GPT | o1-mini | active | | `beta` | 0.9.8 |
| GPT | o1-mini-2024-09-12 | active | | `beta` | 0.9.9 |
| GPT | o1 | active | | vision | `beta` + `untested` | 0.9.10 |
| GPT | o1-2024-12-17 | active | | vision | `beta` + `untested` | 0.9.10 |


`beta` models provide limited control over features or capabilities. Those currently also do not support `system` role: if using `prompt presets` those might not work.

An `untested` model is added but is not able to be tested by the development team until we get access to the model (currently open to [Tier 5](https://platform.openai.com/docs/guides/rate-limits#usage-tiers) customers)

Once a model is `deprecated`, using it in your models list will have it discarded from the available list with a notification.

Similarly, if a used model is listed as `legacy`, a notification of the upcoming deprecation will be shown in the UI.
Expand Down Expand Up @@ -351,7 +356,8 @@ For example, if your `appdata` location for the OpenAI WebUI was `/mnt/user/appd
## 3.2. Version information/Changelog
- v0.9.9 (20241206): API changes to use `o1-mini`
- v0.9.10 (20241217): Added `o1` model (untested) following its API access availability
- v0.9.9 (20241206): API changes to use `o1-mini` and `o1-preview` (tested)
- v0.9.8 (20241010): Added `o1-preview` and `o1-mini` model (untested) + "prompt presets" functionalities
- v0.9.7 (20240718): Added `gpt-4o-mini` and `deprecated` older `32k` models
- v0.9.6 (20240701): Added method to disable `vision` for capable models + added whole WebUI password protection using streamlit's `secrets.toml` method
Expand Down
2 changes: 1 addition & 1 deletion common_functions.py
Original file line number Diff line number Diff line change
Expand Up @@ -10,7 +10,7 @@

from datetime import datetime

iti_version="0.9.9"
iti_version="0.9.10"

def isBlank (myString):
return not (myString and myString.strip())
Expand Down
22 changes: 22 additions & 0 deletions models.json
Original file line number Diff line number Diff line change
Expand Up @@ -115,6 +115,28 @@
"capability": "",
"beta_model": true
},
"o1":
{
"label": "o1: reasoning model designed to solve hard problems across domains. o1 series of large language models are trained with reinforcement learning to perform complex reasoning. o1 models think before they answer, producing a long internal chain of thought before responding to the user. o1 model supports both text and image inputs, and produces text outputs (including Structured Outputs). Points to: o1-2024-12-17",
"max_token": 100000,
"context_token": 200000,
"data": "Unlisted (as of 20241217)",
"status": "active",
"status_details": "beta model: limited control of features and capabilities. Does not support system role. For details, see https://platform.openai.com/docs/guides/reasoning",
"capability": "vision",
"beta_model": true
},
"o1-2024-12-17":
{
"label": "o1: reasoning model designed to solve hard problems across domains. o1 series of large language models are trained with reinforcement learning to perform complex reasoning. o1 models think before they answer, producing a long internal chain of thought before responding to the user.",
"max_token": 100000,
"context_token": 200000,
"data": "Unlisted (as of 20241217)",
"status": "active",
"status_details": "beta model: limited control of features and capabilities. Does not support system role. For details, see https://platform.openai.com/docs/guides/reasoning",
"capability": "vision",
"beta_model": true
},
"gpt-4-turbo":
{
"label": "The latest GPT-4 Turbo model with vision capabilities. Currently points to gpt-4-turbo-2024-04-09.",
Expand Down
4 changes: 2 additions & 2 deletions pyproject.toml
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
[tool.poetry]
name = "Infotrend-OpenAI_WebUI"
version = "0.9.9"
version = "0.9.10"
description = "Streamlit-based WebUI to OpenAI's ChatGPT and Dall-E's API (requires an OpenAI API key)."
authors = ["Martial Michel <[email protected]>"]
license = "MIT"
Expand All @@ -14,7 +14,7 @@ build-backend = "poetry.core.masonry.api"

[tool.poetry.dependencies]
python = "^3.10"
openai = "==1.56.1"
openai = "==1.58.1"
streamlit = "^1.39.0"
extra-streamlit-components = "^0.1.71"
streamlit-extras = "^0.4.7"
Expand Down
6 changes: 3 additions & 3 deletions unraid/OpenAI_WebUI.xml
Original file line number Diff line number Diff line change
Expand Up @@ -39,11 +39,11 @@ Extra parameters are available under the advanced settings.&#xD;
e">/mnt/user/appdata/openai_webui/savedir</Config>
<Config Name="OAIWUI_SAVEDIR" Target="OAIWUI_SAVEDIR" Default="/iti/savedir" Mode="" Description="Required: location where generated files are placed" Type="Variable" Display="advanced" Required="true" Mask="false">/iti/savedir</Config>
<Config Name="OAIWUI_GPT_ONLY" Target="OAIWUI_GPT_ONLY" Default="False" Mode="" Description="If True, only show the GPT UI&#13;&#10;If False, show both the GPT and the DallE UI" Type="Variable" Display="always" Required="true" Mask="false">False</Config>
<Config Name="OAIWUI_GPT_MODELS" Target="OAIWUI_GPT_MODELS" Default="gpt-4,gpt-3.5-turbo" Mode="" Description="Required: comma-separated list of GPT models authorized for the provided API Key" Type="Variable" Display="always" Required="true" Mask="false">gpt-4o,gpt-4-turbo,gpt-4,gpt-3.5-turbo</Config>
<Config Name="OAIWUI_DALLE_MODELS" Target="OAIWUI_DALLE_MODELS" Default="dall-e-3,dall-e-2" Mode="" Description="Required: comma-separated list of DALL-E models authorized for the provided API Key" Type="Variable" Display="always" Required="true" Mask="false">dall-e-3,dall-e-2</Config>
<Config Name="OAIWUI_GPT_MODELS" Target="OAIWUI_GPT_MODELS" Default="gpt-4,gpt-3.5-turbo" Mode="" Description="Required: comma-separated list of GPT models authorized for the provided API Key" Type="Variable" Display="always" Required="true" Mask="false">gpt-4o-mini,gpt-4o,gpt-4-turbo,gpt-4,o1-mini</Config>
<Config Name="OAIWUI_DALLE_MODELS" Target="OAIWUI_DALLE_MODELS" Default="dall-e-3,dall-e-2" Mode="" Description="Required: comma-separated list of DALL-E models authorized for the provided API Key" Type="Variable" Display="always" Required="true" Mask="false">dall-e-3</Config>
<Config Name="OAIWUI_GPT_VISION" Target="OAIWUI_GPT_VISION" Default="True" Mode="" Description="For compatible models, disable their vision capabilities" Type="Variable" Display="always" Required="false" Mask="false">True</Config>
<Config Name="OAIWUI_IGNORE_EMPTY" Target="OAIWUI_IGNORE_EMPTY" Default="" Mode="" Description="Required for Unraid: avoid errors in case some (not all) environment variables are not set" Type="Variable" Display="advanced-hide" Required="true" Mask="false">True</Config>
<Config Name="OAIWUI_USERNAME" Target="OAIWUI_USERNAME" Default="" Mode="" Description="Default username (leave blank to be prompted)" Type="Variable" Display="advanced" Required="false" Mask="false">user2</Config>
<Config Name="OAIWUI_USERNAME" Target="OAIWUI_USERNAME" Default="" Mode="" Description="Default username (leave blank to be prompted)" Type="Variable" Display="advanced" Required="false" Mask="false">unraid_user</Config>
<Config Name="OAIWUI_PROMPT_PRESETS_DIR" Target="OAIWUI_PROMPT_PRESETS_DIR" Default="" Mode="" Description="Directory that contains prompt presets (see GitHub for examples). If a directory is provided, it must contains at least one valid json file" Type="Variable" Display="advanced" Required="false" Mask="false"/>
<Config Name="OAIWUI_PROMPT_PRESETS_ONLY" Target="OAIWUI_PROMPT_PRESETS_ONLY" Default="" Mode="" Description="Json file that contains valid settings to use for the OAIWUI_PROMPT_PRESETS_DIR presets" Type="Variable" Display="advanced" Required="false" Mask="false"/>
</Container>

0 comments on commit 44c721a

Please sign in to comment.