Portkey

We just raised our $15M Series A to scale our unified control plane for production AI. AI is now mission-critical infrastructure, and we’re building the reliability layer so it never breaks. Thanks to our investors @ElevCap + @lightspeedvp portkey.ai/blog/series-a-…

Big news: Portkey has joined @PaloAltoNtwks. Portkey will become the AI Gateway for Prisma AIRS: the control plane for routing, securing, and governing every AI transaction in the enterprise. Following integration, it will offer: • Runtime security on agentic traffic • AI identity + least-privilege access controls • Compliance tooling • 3,000+ LLMs and MCP tools, unified • Production reliability via automated failovers and more Same gateway. Same APIs. Same team. Full story: paloaltonetworks.com/company/press/…

4 point 8. On Portkey.

have you had a peek at our LLMs in Prod leaderboard? some really surprising changes lately!

Who can call which model, and with what parameters, usually ends up as checks scattered across different services and clients. Control it centrally with our new guardrail checks: Model Rules tie model access to request metadata. Allow/deny lives at the gateway instead of being re-implemented in every app. Request-params check validate tools and parameters before the request reaches the LLM.

Running multiple providers and models in production requires more routing control. Here are a few we just added to Portkey: Passthrough targets: if you don't want the gateway to rewrite a specific hop, send it straight to the upstream you name. Sticky sessions: follow-up calls in a conversation will land on the same model instance. Forward headers: copy or rename inbound headers on the way out. Useful if you're passing tracing IDs, tenant markers, or provider-specific headers. Subdomain routing: if your routing logic depends on subdomain variants of your gateway host, Configs now respects that. Read more about this here: portkey.ai/docs/product/a…

Your team probably uses Claude Desktop. But is it governed? You can now route Claude Desktop through Portkey and get: → Route to Anthropic, AWS Bedrock, or Vertex AI from a single endpoint → Per-team budgets, rate limits, and guardrails enforced automatically → Full logs — cost, latency, tokens, metadata — for every prompt Same Claude Desktop experience. See the setup guide here → portkey.ai/docs/integrati…

A great AI Gateway that stays niche is a nice tool. A great AI Gateway that becomes ubiquitous changes how the industry works. Today, @PortkeyAI has signed a definitive agreement to join @PaloAltoNtwks

@retrovrv ❤️

portkey

Exciting News: We've signed a definitive agreement to join Palo Alto Networks! We started Portkey on a single idea: that AI in production would need a control plane. Not a feature, not a category somewhere on the side of the stack. Infrastructure. A place where governance, observability, security, and routing converge so that enterprises can run AI the way they run any system they actually depend on. Three years later, we route trillions of tokens per month for teams running AI at real scale. The idea became the floor. AI control planes don't win by being clever, they win by becoming standard. The default. The thing nobody questions is in the stack. That's the next chapter. Getting there at the speed the moment requires means joining a company with the depth, reach, and enterprise footprint to make the standard real. Palo Alto Networks is that company for us. Link to blog post: paloaltonetworks.com/blog/2026/04/s…

We’re excited to share our intent to acquire @PortkeyAI. Once the acquisition has closed, we plan to integrate Portkey’s AI Gateway into Prisma AIRS to establish a control plane to secure autonomous agents, thereby allowing companies to accelerate AI innovation with confidence. bit.ly/4tEwAZX

Full setup: portkey.ai/docs/integrati…

Right now, every Hermes session needs its own credentials for every MCP server it connects to. Once the agent has access, it has access to everything that server exposes. Portkey's MCP Gateway changes that: - Hermes authenticates once. The gateway handles credential injection for every server behind it - Toggle specific tools on or off at the org or workspace level. Dangerous tools don't surface to the agent at all - Every tool call logged with user, parameters, and full response - Approval workflows for tools that need a human in the loop before they execute - Your Hermes config doesn't change. The agent just calls tools through Portkey's endpoint.

Hermes Agent by @NousResearch now works with Portkey. Here's what that unlocks 🧵

introducing portkey cli. 1 command to set up claude code or codex with portkey's gateway. add llms, mcps, and skills across your entire eng org. every agent gets the same batteries, no manual setup. try it today: npx portkey setup

The Agent Gateway webinar is TODAY. A few hours away, still time to grab a spot. If you're running agents in production you should sign up, now! portkey.sh/agent-webinar

Agent Gateway is in beta. One production layer for every agent in your org. Governance, observability, and access control from a single control plane. Live webinar demo this Friday. Register now: portkey.sh/agent-webinar