Releases: VariantConst/OpenWebUI-Monitor
Releases · VariantConst/OpenWebUI-Monitor
v0.3.8
What's Changed
- Update Next.js to fix Next.js CVE and update most dependecies to lastest minor versions.
- Remove information-leaking log by @LilianBoulard in #95
- feat: Add complete Spanish translation support with i18n enhancements by @TeenBiscuits in #107
- fix: add language selection dropdown to header by @TeenBiscuits in #108
New Contributors
- @LilianBoulard made their first contribution in #95
- @TeenBiscuits made their first contribution in #107
Full Changelog: v0.3.7...v0.3.8
v0.3.7
v0.3.6
v0.3.6-beta.1
v0.3.5
Features
- One-Click Derived Model Pricing Sync: The model page now provides a one-click function to synchronize the pricing of all derived models. Derived models refer to custom models created within the OpenWebUI workspace. This feature ensures that the pricing of derived models remains consistent with their base models. (Implemented #56)
Fixes
- Corrected Usage Statistics for Specific Date Ranges: Fixed an issue where usage statistics within a specified date range were inaccurate. (Fixed #55)
- Clarified Import/Export Button Meaning in Model Page Charts: Improved the clarity of the import/export buttons in the model page.
- Updated Function for More Accurate Token Calculation: Enhanced token calculation accuracy for system prompts and file usage. Thanks to @OrenZhang.
- Synchronized Model Names with OpenWebUI: Model names will now automatically stay in sync with OpenWebUI.
v0.3.4
Fixes
- Corrected Default
per_message_costValue: Fixed an issue where the default value ofper_message_costwas not set to-1, which unintentionally disabled input and output pricing for models. (Fixed #49) - Updated Functions for OpenWebUI v0.5.8+: Ensured compatibility with OpenWebUI v0.5.8 and later versions, thanks to @xinhai-ai @mbeltagy .
- Accurate Token Consumption Calculation: System prompts and file token usage are now correctly accounted for, thanks to @xinhai-ai .
v0.3.3
Fixes
- Model Retrieval Error in v0.3.2: Fixed an issue where retrieving models caused errors in version v0.3.2.
- Improved Model Loading Speed: Optimized the model loading process for faster performance.
- Negative Balance and Pricing Display: When user balance or model pricing is negative, it will no longer display as "Not Set" (except for per-message pricing).
v0.3.2
Features
- New
COST_ON_INLETEnvironment Variable: IntroducedCOST_ON_INLETto penalize users who exit the page before completing a response, preventing fee evasion. When enabled, a deduction is made from the balance immediately after a request is initiated and refunded upon completion of the conversation. This can be set globally or for specific model IDs. If the model is unavailable, the deducted amount will not be refunded, so use this option cautiously. (Fixed #27) - Automatic Base Model Price Inheritance: Models created in the "Workspace" will now automatically inherit the pricing of their base model instead of using a default value. (Implemented #32)
Fixes
- Accurate Total Consumption Display: Fixed an issue where the total consumption amount was displayed incorrectly. (Fixed #38)
- No Charges for Failed Model Requests: Ensured that no fees are deducted when model requests fail. (Fixed #33)
- Functions Update for OpenWebUI v0.5.8+: Updated functions to ensure compatibility with OpenWebUI v0.5.10 and above. (Fixed #39)
v0.3.1
Fixes
- User List Retrieval Issue: Fixed an issue where the user list failed to load correctly.
- Time Range Selection in Statistics Page: Resolved an issue preventing proper selection of time ranges on the statistics page.
v0.3.0
Features
- Comprehensive UI/UX Upgrade: Redesigned the entire interface for a modern look and improved user experience.
- New Reporting Capabilities: Introduced daily, weekly, and monthly report generation for detailed usage insights.
- User Blacklist Functionality: Added the ability to blacklist users, preventing service access when their balance is insufficient. This update also addresses issue #26.
Fixes
- Multilingual Support Enhancement: Enhanced multilingual capabilities, setting English as the default language for error messages (Fixed #29).
- Documentation Updates: Improved and updated relevant documentation for better clarity and usability (Fixed #25).
- User Balance Restrictions: Implemented stricter controls on balance modifications to ensure data integrity (Fixed #22).
- Precision in Financial Calculations: Adjusted model pricing and user balance calculations to support precision up to four decimal places (Fixed #16).