> For the complete documentation index, see [llms.txt](https://docs.tolvyn.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.tolvyn.io/migrate-to-tolvyn/from-portkey.md).

# From Portkey

Two things change: the base URL and how you authenticate. Everything else stays the same.

***

## What changes

|              | Portkey                               | TOLVYN                                               |
| ------------ | ------------------------------------- | ---------------------------------------------------- |
| Base URL     | `https://api.portkey.ai/v1`           | `https://proxy.tolvyn.io/v1/proxy/openai/v1`         |
| Auth         | `x-portkey-api-key: pk_...`           | `Authorization: Bearer tlv_live_...`                 |
| Provider key | Virtual key (`x-portkey-virtual-key`) | Stored encrypted server-side (set once in dashboard) |

Your prompts, models, parameters, and response format are unchanged.

***

## Before / after

### Python (openai SDK)

**Before:**

```python
from openai import OpenAI

client = OpenAI(
    api_key="anything",  # ignored by Portkey
    base_url="https://api.portkey.ai/v1",
    default_headers={
        "x-portkey-api-key": "pk_...",
        "x-portkey-virtual-key": "openai-vk-...",
        "x-portkey-trace-id": "trace-abc123",
    },
)
```

**After:**

```python
from openai import OpenAI

client = OpenAI(
    api_key="tlv_live_...",   # your TOLVYN key
    base_url="https://proxy.tolvyn.io/v1/proxy/openai/v1",
)
```

Add your real OpenAI key once in the TOLVYN dashboard on the **Providers** page (`/providers`) — it is stored encrypted server-side and never sent to the client.

***

### Node.js (openai SDK)

**Before:**

```js
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "anything",
  baseURL: "https://api.portkey.ai/v1",
  defaultHeaders: {
    "x-portkey-api-key": "pk_...",
    "x-portkey-virtual-key": "openai-vk-...",
  },
});
```

**After:**

```js
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "tlv_live_...",
  baseURL: "https://proxy.tolvyn.io/v1/proxy/openai/v1",
});
```

***

### curl

**Before:**

```bash
curl https://api.portkey.ai/v1/chat/completions \
  -H "x-portkey-api-key: pk_..." \
  -H "x-portkey-virtual-key: openai-vk-..." \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-4o-mini","messages":[{"role":"user","content":"Hello"}]}'
```

**After:**

```bash
curl https://proxy.tolvyn.io/v1/proxy/openai/v1/chat/completions \
  -H "Authorization: Bearer tlv_live_..." \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-4o-mini","messages":[{"role":"user","content":"Hello"}]}'
```

***

## Header mapping

TOLVYN reads Portkey metadata headers automatically during migration.

| Portkey Header          | TOLVYN Equivalent                    | Notes                                            |
| ----------------------- | ------------------------------------ | ------------------------------------------------ |
| `x-portkey-api-key`     | `Authorization: Bearer tlv_live_...` | Replace with your TOLVYN key                     |
| `x-portkey-virtual-key` | Provider key stored server-side      | Set once in TOLVYN dashboard; remove from client |
| `x-portkey-trace-id`    | stored in `tags.portkey_trace_id`    | Auto-captured                                    |
| `x-portkey-metadata`    | merged into `tags.portkey_<key>`     | JSON object values auto-captured                 |

All Portkey headers are stripped before the request reaches OpenAI — they will not appear in provider logs.

***

## Virtual keys vs. server-side provider keys

Portkey uses virtual keys to reference stored credentials from the client. TOLVYN stores the real provider key encrypted on the server — it never leaves.

1. Go to **Providers** (`/providers`) in the TOLVYN dashboard.
2. Add your OpenAI (or Anthropic / Google) key.
3. Remove the `x-portkey-virtual-key` header from your client code.

Your client never holds a real provider key again.

***

## Feature comparison

| Feature                     | Portkey | TOLVYN |
| --------------------------- | ------- | ------ |
| Request logging             | ✅       | ✅      |
| Cost tracking               | ✅       | ✅      |
| Multi-provider routing      | ✅       | ✅      |
| Budget enforcement          | limited | ✅      |
| Hard kill switch            | ❌       | ✅      |
| Tamper-evident audit ledger | ❌       | ✅      |
| Per-customer attribution    | ❌       | ✅      |
| Invoice reconciliation      | ❌       | ✅      |
| Savings analysis            | ❌       | ✅      |

**On pricing:** Portkey Pro is $49/mo at the time of writing, with basic observability and routing. TOLVYN Growth is ₹17,999/month (approx. $199 USD) with budget enforcement, kill switches, a financial-grade audit trail, per-customer attribution, and invoice reconciliation. If you only need routing and logs, Portkey is cheaper. If you need financial control and accountability, TOLVYN is built for that.

***

## TOLVYN-native headers (optional)

Once migrated, adopt TOLVYN's attribution headers for richer breakdowns:

```
X-Tolvyn-Team: backend-team
X-Tolvyn-Service: summarization
X-Tolvyn-Feature: report-gen
X-Tolvyn-User: user-123
X-Tolvyn-End-Customer: acme-corp
```

These are all optional. The dashboard shows cost and usage breakdowns per team, service, user, and end customer automatically.