May
May-king it production ready✨
In May, we shipped the kind of upgrades that help you move your AI Agents fast into productiion and stay in control — whether you’re scaling, securing AI behavior, or bringing new models to your apps.
We launched deep integrations with agent frameworks like PydanticAI and OpenAI Agents SDK, added enterprise-grade controls to Claude Code, made it simpler to call a remote MCP server simpler and much more!
Here’s everything new this month:
Tracing Integrations: Arize AI
For teams consolidating observability into Arize, you can now view Portkey’s logs directly into Arize Phoenix to get unified trace views across your LLM workflows.
With Portkey’s deep integration into the Azure AI ecosystem (OpenAI, Foundry, APIM, Marketplace), teams can now build, scale, and govern GenAI apps without leaving their existing cloud setup.
Our customers are vouching for it!
You can now assign multiple labels to a single prompt version, making it easy to promote a version across environments like staging and production.
Gateway to any API
Portkey now supports 
Summary
| Area | Key Highlights |
|---|---|
| Platform | • Full HTTP method support (GET, PUT, DELETE)• OTel analytics export to your stack • OpenAI Computer Use Tool support • Multimodal embedding support (Vertex AI) |
| Enterprise | • Deep Azure AI ecosystem integration (Foundry, APIM, Marketplace) • Claude Code with enterprise controls (rate limits, observability) • Model whitelist guardrail for org/env control |
| Integrations | • Expanded AI Agent Frameworks (PydanticAI, OpenAI SDK, Strands) • Support for latest models (Claude 4, Grok 3, Gemini 2.5) & new providers • AI Coding Assistant integrations (Cline, Roo Code) • Remote MCP server & Arize Phoenix tracing support |
| Security | • New Prompt Security guardrails (injection, data protection) • JWT validator input guardrail • PANW Prisma AIRS plugin for real-time risk blocking |
| Resources | • New Solution Pages (AWS Bedrock, GovCloud) • New Cookbooks (OpenAI Computer Use, Llama Prompt Ops) |
AI Agent Infrastructure
AI agent frameworks are helping teams prototype faster, but taking agents to production requires real infrastructure. Portkey integrates with leading frameworks to bring interoperability, observability, reliability, and cost management to your agent workflows.PydanticAI
PydanticAI is a Python framework that brings FastAPI-like ergonomics to building AI agents.
OpenAI Agents SDK
OpenAI Agents SDK helps teams ship production-grade agents with built-in planning, memory, and tool use.
Strands Agents
Strands Agents is a lightweight agent framework built by AWS to simplify agent development.
Remote MCP servers
Portkey now supports calling a remote MCP server that is maintained by developers and organizations across the internet that expose these tools to MCP clients via the Responses API Read more about the integration here.Azure AI ecosystem
More than half of Fortune 500 companies use Azure OpenAI. But building GenAI apps in the enterprise is still messy, cost attribution, routing logic, usage tracking, model evaluation… all scattered.Portkey for AI Tools
Claude Code
Bring enterprise-grade visibility, governance, and access control to Claude Code.
Cline
Supercharge your AI-powered terminal with cost tracking, access controls, and observability.
Roo Code
Add security, compliance, and real-time analytics to your code assistant workflows.
Goose
Add essential enterprise controls to Goose’s powerful autonomous coding capabilities
Multilmodal embeddings
Portkey now supports embedding APIs from Vertex AI for text, image, and video—across multiple languages. This unlocks the ability to:- Build multimodal search and retrieval
- Power multimodal RAG pipelines
- Track, route, and optimize embedding usage at scale
Platform
Multi-label support for promptsGET, PUT, and DELETE HTTP methods in addition to POST, allowing you to route requests to any external or self-hosted provider endpoint. This means you can connect to custom APIs directly through Portkey with full observability for every call.
OTel Integration (Analytics Data)
You can now export Portkey analytics to any OpenTelemetry (OTel)-compatible collector, integrating easily into your existing observability stack.
Improvements
- Token cost tracking is now available for
gpt-image-1. - Ping messages are removed from streamed responses.
- Resizing metadata columns in logs
New Models & Providers
- New additions
- Claude 4 is now live for advanced reasoning and coding.
- Grok 3 & Grok 3 Mini are available on Azure
- Lepton AI is now live
- Nscale Models can now be accessed through Portkey.
- Updates
- PDF Support for Claude via Anthropic and Bedrock.
- Gemini 2.5 Thinking Mode is now supported in Prompt Playground.
- Extended Thinking is available for Claude 3.7 and Claude 4.
- Image generation now supported on WorkersAI
- Tool Calling and Function Calling for Mistral is now live.
- MIME Type is now supported for Vertex AI
Guardrails
- Prompt Security guardrails: Integrate with Prompt Security to detect prompt injection and prevent sensitive data exposure in both prompts and responses.
- JWT validator guardrail: Added as an input guardrail to validate incoming JWT tokens before requests are sent to the LLM.
- PANW Prisma AIRS Plugin: Portkey now integrates with Palo Alto Networks’ AIRS (AI Runtime Security) to enforce guardrails that block risky prompts or model responses based on real-time security analysis.
- Model whitelist guardrail: Restrict or deny specific models at the org, environment, or request level using a flexible whitelist/blacklist guardrail.
Resources
- Cookbook: Optimizing Prompts with LLama Prompt Ops
- Cookbook: OpenAI Computer Use Tool
- Guardrail documentation is now located under “Integrations”.
- Expanded guides for agent frameworks, including CrewAI and LangGraph.