> ## Documentation Index
> Fetch the complete documentation index at: https://docs.portkey.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# February

We raised our \$15M Series A this month! 🎉

Led by Elevation Capital with participation from Lightspeed, this funding helps us double down on our mission: building AI infrastructure that never breaks.

We've also made improvements across the gateway, guardrails, and provider ecosystem this month.

See what's new!

## Summary

| Area                     | Key highlights                                                                                                                                              |
| ------------------------ | ----------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Platform**             | Model Catalog migration for all orgs                                                                                                                        |
| **Gateway**              | Use Responses API and Messages API with any provider, gRPC support, JWT default configs, Portkey skills for AI coding assistants                            |
| **Guardrails**           | Protect your apps with Zscaler AI Guard                                                                                                                     |
| **Models and providers** | Route to Databricks, use Claude 4.6 features, run Together AI reasoning models, Veo video generation on Vertex AI, Passthrough provider, stream TTS via SSE |
| **Community & Events**   | Agent Harness Salon: BLR (Sat 22 Feb), RSA Conference SF (March)                                                                                            |

## Highlights

We're thrilled to announce our \$15M Series A! This is a huge milestone for Portkey and a testament to the incredible trust our customers and community have placed in us. With this funding, we're doubling down on our mission: building the unified control plane for production AI that never breaks.

<img src="https://userimg-assets.customeriomail.com/images/client-env-144095/1771507281211_SM%20Post%20Large_01KHV0ZVGGKCMQK72HKYC0QXF8.jpeg" alt="Portkey Series A announcement graphic" style={{ maxWidth: "80%", borderRadius: "8px", margin: "24px 0" }} />

Here's what we're going to focus on:

* **Expanding go-to-market** — Meeting the growing enterprise demand across finance, pharma, technology, and beyond
* **Governance for agentic AI** — Building the controls organizations need as agents take autonomous action: permissions, identity, access boundaries, and budget guardrails
* **Platform infrastructure at scale** — Higher-volume workloads, real-time use cases, and day0 support for new models and pricing changes

[Read the full announcement here](https://portkey.ai/blog/series-a-funding/)

## Platform

### All Organizations Now on Model Catalog

We've upgraded all organizations to Model Catalog. It gives you a unified way to discover, configure, and route to models across providers.

* Browse all available models across 40+ providers in one place
* Configure model-specific settings without touching code
* Switch providers for the same model with a single change

[Explore Model Catalog](/product/model-catalog)

## Gateway

### Use the Responses API with Any Provider

You can now use OpenAI's Responses API (`/v1/responses`) across providers!

* Keep a single API format while switching between Anthropic, Google, Bedrock, and others
* Use prompt caching and thinking parameters across providers

```python theme={"system"}
from portkey_ai import Portkey

portkey = Portkey(api_key="PORTKEY_API_KEY")

response = portkey.responses.create(
    model="@anthropic-provider/claude-sonnet-4-5-20250514",
    input="Explain quantum computing in simple terms"
)

print(response.output_text)
```

[Get started with Responses API](/product/ai-gateway/responses-api)

### Use the Messages API with Any Provider

You can now use Anthropic's Messages API (`/v1/messages`) with any provider through a universal adapter — not just Anthropic, Bedrock, and Vertex AI.

```python theme={"system"}
import anthropic

client = anthropic.Anthropic(
    api_key="dummy", # we will use portkey's provider slug
    default_headers={"x-portkey-api-key": "YOUR_PORTKEY_API_KEY"},
    base_url="https://api.portkey.ai"
)

message = client.messages.create(
    model="@google-provider/gemini-2.5-flash",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Explain quantum computing in simple terms"}]
)

print(message.content[0].text)
```

* Keep your existing Messages API code while routing to OpenAI, Google, and more
* Let the gateway handle format conversion automatically

See how to use the [Messages endpoint](http://localhost:3000/product/ai-gateway/messages-api)

### gRPC Support

The Portkey Gateway now supports gRPC as an alternative transport protocol alongside HTTP/REST. This enables lower latency, efficient binary serialization via Protocol Buffers, and native streaming support for applications that prefer gRPC communication.

### Embed Default Configs in JWT Tokens

You can now embed a default config directly in the JWT payload using the `defaults` claim. This works the same way as attaching default configs to API keys, but lets you control it per-token — useful when different users or services need different routing rules.

### Portkey Skills for AI Coding Assistants

Give your AI coding assistant the knowledge to work with Portkey SDKs. Install our official skills to teach Cursor, Claude Code, GitHub Copilot, and other assistants how to use Portkey's APIs, patterns, and best practices.

```bash theme={"system"}
npx add-skill portkey-ai/skills
```

[Set up Portkey skills for your AI assistant](/api-reference/inference-api/agentic-usage)

## Guardrails

### Protect Your Apps with Zscaler AI Guard

You can now connect Zscaler AI Guard to scan prompts and responses for security threats.

* Enforce Detection Policies for security checks
* Block or flag data loss risks with DLP protection
* Catch prompt injection attempts on both inputs and outputs

[Connect Zscaler AI Guard](/integrations/guardrails/zscaler)

## Models and providers

<ul>
  <li>
    <b>Databricks</b>: You can now route requests to Databricks Model Serving
    for chat completions, completions, and embeddings.{" "}
    <a href="/integrations/llms/databricks">Set up Databricks</a>.
  </li>

  <li>
    <b>Claude 4.6</b>: Use Claude 4.6 features across Anthropic, Bedrock, and
    other providers — including Adaptive Thinking with{" "}
    <code>reasoning\_effort</code>, structured outputs via{" "}
    <code>output\_config</code>, and new stop reasons like <code>refusal</code>.
  </li>

  <li>
    <b>Together AI reasoning</b>: Run reasoning/thinking models on Together AI
    with the <code>reasoning\_effort</code> parameter and get structured{" "}
    <code>content\_blocks</code> in responses.{" "}

    <a href="/integrations/llms/together-ai#reasoning-thinking-support">
      Try Together AI reasoning
    </a>

    .
  </li>

  <li>
    <b>Bedrock Anthropic citations</b>: Access Anthropic's citations feature on
    Bedrock through the chat completions API.
  </li>

  <li>
    <b>Veo on Vertex AI</b>: Generate videos from text or image prompts using
    Google's Veo model. Uses a two-step long-running flow: submit a generation
    request, receive an operation name, then poll until the video is ready.
  </li>

  <li>
    <b>Passthrough provider</b>: Send requests exactly as they are to any
    backend provider, without Portkey's usual parameter mapping or validation.
    Useful when you need full control over the request body or when working with
    provider-specific features.
  </li>
</ul>

### Enhancements

<ul>
  <li>
    <b>OpenAI & Azure OpenAI TTS streaming</b>: Stream text-to-speech audio via
    Server-Sent Events by setting <code>stream\_format: "sse"</code>.{" "}

    <a href="/product/ai-gateway/multimodal-capabilities/text-to-speech#sse-streaming">
      Set up SSE streaming
    </a>

    .
  </li>

  <li>
    <b>ZhipuAI</b>: Generate images with ZhipuAI's CogView models (e.g.,{" "}
    <code>cogview-4-250304</code>).{" "}
    <a href="/integrations/llms/zhipu#image-generation">See ZhipuAI docs</a>.
  </li>

  <li>
    <b>Vertex AI</b>: Control image and video input resolution with{" "}
    <code>media\_resolution</code>, skip PTU cost attribution with{" "}
    <code>vertex\_skip\_ptu\_cost\_attribution</code>, and configure workload
    identity auth via <code>x-portkey-vertex-auth-type</code>.
  </li>

  <li>
    <b>Batch pricing</b>: Get accurate cost attribution for batch requests with
    dedicated batch pricing. When batch-specific pricing isn't available, costs
    default to 50% of standard pricing.
  </li>
</ul>

## Community & Events

### Agent Harness Salon: Bangalore

We hosted another Agent Harness Salon in Bangalore on Saturday, February 22nd! Thanks to everyone who joined for the demos, discussions, and drinks.

<img src="https://mintcdn.com/portkey-docs/Mc5UPhriqhb31md1/images/changelog/Agentharnesses.png?fit=max&auto=format&n=Mc5UPhriqhb31md1&q=85&s=3250546a1ec8f605835ded712f59eefd" alt="Agent Harness Salon Bangalore February 2026" style={{ maxWidth: "80%", borderRadius: "8px", margin: "24px 0" }} width="1179" height="808" data-path="images/changelog/Agentharnesses.png" />

### Meet us at RSA!

The Portkey team will be in SF for RSA Conference this month! If you're attending and want to chat about AI security, governance, or infrastructure, we'd love to connect.

[Book a slot here](https://portkey.sh/rsa)

## Resources

* **Blog:** [LLM Deployment Pipeline Explained Step by Step](https://portkey.ai/blog/llm-deployment/)
* **Blog:** [How to host an AI Hackathon without losing control of your keys or budget](https://portkey.ai/blog/how-to-host-an-ai-hackathon-without-losing-control-of-your-keys-or-budget/)
* **Blog:** [Securing the MCP Gateway: Lasso Partners with Portkey to Deliver Enterprise-Grade Agentic AI Protection](https://portkey.ai/blog/securing-mcp-to-deliver-enterprise-grade-agentic-ai-protection/)
* **Blog:** [The best approach to compare LLM outputs](https://portkey.ai/blog/the-best-approach-to-compare-llm-outputs/)

## Support

<CardGroup cols={2}>
  <Card title="Need Help?" icon="bug" href="https://github.com/Portkey-AI/gateway/issues">
    Open an issue on GitHub
  </Card>

  <Card title="Join Us" icon="discord" href="https://portkey.wiki/community">
    Get support in our Discord
  </Card>
</CardGroup>