Schedule Call
Discuss how Portkey’s AI Gateway can enhance your organization’s AI infrastructure
v1.5.0
Fixes and Improvements
Needs a Helm repo update(>app-1.4.0)
- Security patch — removed root user from container image (BREAKING CHANGE). The container image used by this chart no longer runs as root. The image now runs processes as a non-root UID and enforces a non-root container securityContext. Requires Helm repo upgrade (>app-1.4.0) to deploy the new image and chart settings.
- New Security Feature to provide access to Workspace Members to Manage Configs
- UI update to highlight Traces by their category
- UI update to improve visibility of Logs with Guardrails
- Update zhipu provider to include new models
- Improve the accessibility of Login/Signup pages
v1.4.9
Fixes and Improvements
- Added a JSON mode for prompt parameters, allowing real-time updates and easier editing.
- Improvement of MCP Consent page to search by workspaces
- Fix for Model Catalog page to display the Usage Limit on page load
- Fix for Prompt Labels delete functionality. This will remove the label from the prompt version which it is assigned to before deletion.
- Analytics Filter API key now have search functionality.
- Support added for transcribe models under azure and OpenAI providers.
- Added more detailed support for Logs with Tools. On hover fields, description and their types will be visible.
- Configuration options for Contact Us, Discord, and custom support are now available.
- Added support for new OpenAI Models.
v1.4.6
API Key Enhancements
- Added multiple rate limit support. It is now possible to set both cost-based and token-based rate limits on a single key
- Introduced a security setting to restrict the maximum number of
UserAPI keys that can be created per user - Introduced a security setting to configure the default and maximum expiry of
UserAPI keys
MCP Improvements
- Updated the MCP OAuth consent screen to allow selection from eligible workspaces
- Added better states for consent approval, rejection, and failures
- Fixed an issue to allow Workspace admins to edit Workspace-created MCP integrations
Fixes and Improvements
- Hide embedding models in prompt playground
- Allow updating base models for custom models
- Made alias a required field for Azure-OpenAI deployment configurations
- Added gpt-5-codex, gpt-5-pro, and several other OpenAI and Azure-OpenAI models
- Added new Together AI and Grok models
- Added AWS Bedrock Titan Embedding and Titan v1 models
v1.4.0
MCP Enhancements
- Added enhancements for the MCP integrations and servers management
- Updates to the MCP OAuth consent screen
Fixes and Improvements
- Added span name as tooltip for better trace visibility
- Fixed Vertex AI Anthropic model name mapping issue
- Fixed Azure-OpenAI
Add Modelbutton state in update view - Removed config, metadata, and rate limits fields for admin keys
- Fixed display base_url for Model Catalog Python SDK code snippets