Portkey

BYOK mode
Config-driven

# Before
from openai import OpenAI
import os

client = OpenAI(
    base_url="https://api.portkey.ai/v1",
    api_key=os.environ["OPENAI_API_KEY"],
    default_headers={
        "x-portkey-api-key": os.environ["PORTKEY_API_KEY"],
        "x-portkey-provider": "openai",
    },
)

# After
from openai import OpenAI
import os

client = OpenAI(
    base_url="https://api.concentrate.ai/v1",
    api_key=os.environ["CONCENTRATE_API_KEY"],
)

// Before
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.portkey.ai/v1",
  apiKey: process.env.OPENAI_API_KEY,
  defaultHeaders: {
    "x-portkey-api-key": process.env.PORTKEY_API_KEY,
    "x-portkey-provider": "openai",
  },
});

// After
const client = new OpenAI({
  baseURL: "https://api.concentrate.ai/v1",
  apiKey: process.env.CONCENTRATE_API_KEY,
});

# Before
curl https://api.portkey.ai/v1/chat/completions \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "x-portkey-api-key: $PORTKEY_API_KEY" \
  -H "x-portkey-provider: openai" \
  -H "Content-Type: application/json" \
  -d '{ "model": "gpt-4o", "messages": [...] }'

# After
curl https://api.concentrate.ai/v1/chat/completions \
  -H "Authorization: Bearer $CONCENTRATE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "model": "gpt-4o", "messages": [...] }'

# Before
from openai import OpenAI
import os

client = OpenAI(
    base_url="https://api.portkey.ai/v1",
    api_key=os.environ["OPENAI_API_KEY"],
    default_headers={
        "x-portkey-api-key": os.environ["PORTKEY_API_KEY"],
        "x-portkey-config": "pc-prod-fallback",
    },
)

# After
from openai import OpenAI
import os

client = OpenAI(
    base_url="https://api.concentrate.ai/v1",
    api_key=os.environ["CONCENTRATE_API_KEY"],
)
# Fallbacks / caching / retries move from the Config object to body params
# on each request. See Step 4 for the mapping.

// Before
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.portkey.ai/v1",
  apiKey: process.env.OPENAI_API_KEY,
  defaultHeaders: {
    "x-portkey-api-key": process.env.PORTKEY_API_KEY,
    "x-portkey-config": "pc-prod-fallback",
  },
});

// After
const client = new OpenAI({
  baseURL: "https://api.concentrate.ai/v1",
  apiKey: process.env.CONCENTRATE_API_KEY,
});
// Fallbacks / caching / retries move from the Config object to body params
// on each request. See Step 4 for the mapping.

# Before
curl https://api.portkey.ai/v1/chat/completions \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "x-portkey-api-key: $PORTKEY_API_KEY" \
  -H "x-portkey-config: pc-prod-fallback" \
  -H "Content-Type: application/json" \
  -d '{ "model": "gpt-4o", "messages": [...] }'

# After
curl https://api.concentrate.ai/v1/chat/completions \
  -H "Authorization: Bearer $CONCENTRATE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "model": "gpt-4o", "messages": [...], "routing": { "model": { "fallbacks": ["..."] } } }'

Prerequisites

A Concentrate AI account with an active API key

An existing Portkey integration

This guide assumes you are calling Portkey from the OpenAI SDK, the portkey-ai SDK, fetch, requests, or another HTTP client (in BYOK or Config-driven mode).

Quick Start for Claude Code users

If you use Claude Code, you can install a skill that walks through this migration interactively. It searches your project for Portkey usage, strips x-portkey-* headers, decomposes Configs, maps model slugs, and generates a verification script. Drop the skill into your ~/.claude/skills/ directory:

mkdir -p ~/.claude/skills/migrate-portkey && \
  curl -fsSL https://concentrate.ai/scripts/migrate-portkey.md \
  -o ~/.claude/skills/migrate-portkey/SKILL.md

Then start a Claude Code session in your project and ask it to “migrate from Portkey to Concentrate” or run /migrate-portkey. Claude will load the skill and run the steps.

Step 1: Update Your Environment Variables

Replace your Portkey key and any upstream provider keys with a single Concentrate key:

# Before (BYOK mode)
export PORTKEY_API_KEY="pk-..."
export OPENAI_API_KEY="sk-..."
export BASE_URL="https://api.portkey.ai/v1"

# After
export CONCENTRATE_API_KEY="sk-cn-v1-..."
export BASE_URL="https://api.concentrate.ai/v1"

You no longer need OPENAI_API_KEY, ANTHROPIC_API_KEY, AWS credentials, Azure deployment IDs, or any other upstream provider key. Concentrate owns those credentials. Comment them out (don’t delete) until you’ve verified the migration end-to-end, then remove them.

Step 2: Update Your Client

The two big changes are the base URL and stripping every x-portkey-* header. If you were using the portkey-ai SDK, swap to the OpenAI SDK pointed at Concentrate. Concentrate does not ship a dedicated SDK because the OpenAI-compatible shape covers every endpoint.

# Before (Portkey with portkey-ai SDK)
from portkey_ai import Portkey

client = Portkey(
    api_key=os.environ["PORTKEY_API_KEY"],
    virtual_key="openai-prod",
    trace_id="req_42",
    metadata={"_user": "user_42", "feature": "summarizer"},
)

# After (Concentrate)
from openai import OpenAI
import os

client = OpenAI(
    base_url="https://api.concentrate.ai/v1",
    api_key=os.environ["CONCENTRATE_API_KEY"],
)

response = client.chat.completions.create(
    model="anthropic/claude-opus-4-6",
    messages=[{"role": "user", "content": "Say hello in one word"}],
)
print(response.choices[0].message.content)

// Before (Portkey with portkey-ai SDK)
import Portkey from "portkey-ai";

const client = new Portkey({
  apiKey: process.env.PORTKEY_API_KEY,
  virtualKey: "openai-prod",
  traceID: "req_42",
  metadata: { _user: "user_42", feature: "summarizer" },
});

// After (Concentrate)
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.concentrate.ai/v1",
  apiKey: process.env.CONCENTRATE_API_KEY,
});

const response = await client.chat.completions.create({
  model: "anthropic/claude-opus-4-6",
  messages: [{ role: "user", content: "Say hello in one word" }],
});
console.log(response.choices[0].message.content);

curl https://api.concentrate.ai/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $CONCENTRATE_API_KEY" \
  -d '{
    "model": "anthropic/claude-opus-4-6",
    "messages": [{"role": "user", "content": "Say hello in one word"}]
  }'

Step 3: Remove `x-portkey-*` Headers

None of Portkey’s custom headers carry over to Concentrate, so they should come out. They’re dead weight in your client config and mislead future readers. Expand the tables below for the full mapping if any of these are in your code.

Request header mapping

Portkey header	Concentrate replacement
`x-portkey-api-key`	Standard `Authorization: Bearer sk-cn-v1-...`
`x-portkey-provider`, `x-portkey-virtual-key`	Drop. Provider is encoded in the model slug (`provider/model-id`); credentials are managed by Concentrate
`x-portkey-config`	Drop. See Step 4 for per-feature mapping
`x-portkey-trace-id`	Optional `X-Request-Id` for correlation. Concentrate returns its own request id on the response
`x-portkey-metadata`	Drop. Use per-user, per-team, or per-org keys; analytics roll up by key automatically
`x-portkey-cache-force-refresh`	No equivalent. Caching is provider-native, not gateway-stored
`x-portkey-cache-namespace`	`prompt_cache_key` body param
`x-portkey-request-timeout`	Set via your HTTP client (e.g. OpenAI SDK `timeout`)
`x-portkey-forward-headers`, `x-portkey-sensitive-headers`	Not applicable. No BYOK passthrough
`x-portkey-custom-host`	Not supported. Self-hosted models stay on direct calls
`x-portkey-azure-*`	Drop. Request `azure/<model-id>`
`x-portkey-vertex-*`	Drop. Request `ai-studio/<model-id>` or the appropriate Vertex-backed slug
`x-portkey-aws-*`	Drop. Request `bedrock/<model-id>`

`portkey-ai` SDK parameter equivalents

If your code uses the portkey-ai SDK, the snake_case (Python) or camelCase (Node) parameters map to the same headers above:

`portkey-ai` parameter	Header it sets	Concentrate replacement
`api_key` / `apiKey`	`x-portkey-api-key`	OpenAI SDK `api_key` / `apiKey` with your `sk-cn-v1-` key
`virtual_key` / `virtualKey`	`x-portkey-virtual-key`	Drop. Managed credentials
`config`	`x-portkey-config`	Drop. Express settings as body params
`provider`	`x-portkey-provider`	Drop. Model slug carries this
`trace_id` / `traceID`	`x-portkey-trace-id`	Drop or send `X-Request-Id`
`metadata`	`x-portkey-metadata`	Drop. Use per-key/team/org rollups
`cache_force_refresh` / `cacheForceRefresh`	`x-portkey-cache-force-refresh`	Drop
`cache_namespace` / `cacheNamespace`	`x-portkey-cache-namespace`	`prompt_cache_key` body param
`custom_host` / `customHost`	`x-portkey-custom-host`	Not supported
`forward_headers` / `forwardHeaders`	`x-portkey-forward-headers`	Not supported

Response header mapping

Portkey echoes a handful of headers back on every response. If your code reads or logs them, here’s the mapping:

Portkey response header	Concentrate equivalent
`x-portkey-trace-id`	`X-Request-Id` (surfaced in dashboard logs)
`x-portkey-cache-status`	Read `usage.prompt_tokens_details.cached_tokens` in the response body (where supported)
`x-portkey-provider`, `x-portkey-last-used-option-index`	No header. Resolved provider is recorded per request in the dashboard
Rate-limit headers	Standard `X-RateLimit-*` headers. See Errors for 429 semantics

Step 4: Decompose Your Portkey Config

Portkey’s x-portkey-config bundles caching, fallbacks, load balancing, retries, conditional routing, and timeouts into a single saved or inline object. Concentrate does not have a Config primitive. Each behavior is either on by default or expressed as a body param on the request. Expand the table below for the full mapping.

Config key mapping

Portkey config key	Concentrate equivalent
`strategy: { mode: "fallback" }` + `targets`	`routing.model.fallbacks` or `routing.provider.fallbacks`, or `model: "auto"`
`strategy: { mode: "loadbalance" }` + weighted `targets`	`model: "auto"` with `routing.model.sort` across eligible providers
`strategy: { mode: "conditional" }` (route by metadata)	Capability-driven only. Issue different keys per condition and pick in application code
`cache: { mode: "simple" \| "semantic", max_age }`	Provider-native prompt caching. No semantic cache
`retry: { attempts, on_status_codes }`	Automatic failover on any provider error across your fallback chain
`request_timeout`	Set via your HTTP client
`override_params`	Pass directly in the request body
`virtual_key` inside a target	Drop. Managed credentials
`custom_host` inside a target	Not supported
`guardrails`	Guardrails & Redaction

Concentrate has conditional routing, but it conditions on request capabilities (feature support, ZDR flag, per-feature uptime, cache affinity), not on request attributes (metadata, headers, user cohorts). If your Portkey Config routes on _environment, _user, or custom metadata, the migration path is to issue separate keys per condition and pick the key in application code.

Step 5: Update Model Identifiers

Concentrate accepts model strings in two forms:

Bare slug, e.g. gpt-4o, claude-haiku-4-5, auto. Routing picks a provider.
provider/model-id, e.g. bedrock/claude-haiku-4-5, openai/gpt-4o. Pins the request to a specific provider.

If you were on Portkey with model strings like gpt-4o or claude-sonnet-4-6 and provider selection via x-portkey-provider or a Virtual Key, drop the header and let the model slug carry provider intent. Most Portkey model strings work as-is once you remove the provider header. One thing to know about the slashed form: the prefix is the provider that serves the request, not the model’s author. Portkey customers steering through x-portkey-provider: bedrock (or a Bedrock Virtual Key) with model: "claude-sonnet-4-6" should switch to model: "bedrock/claude-haiku-4-5" (or whichever slug GET /v1/models returns). For most popular names the two are the same string (openai, anthropic, mistral), but they diverge whenever a model is hosted by something other than its author:

Author	Provider serving the request	Concentrate slug
Anthropic (`claude-haiku-4-5`)	Anthropic	`anthropic/claude-haiku-4-5`
Anthropic (same model, different host)	AWS Bedrock	`bedrock/claude-haiku-4-5`
Anthropic (same model, different host)	Azure	`azure/claude-haiku-4-5`
Google (`gemini-3.5-flash`)	Google AI Studio	`ai-studio/gemini-3.5-flash`
Meta (`llama-3-8b-instruct`)	AWS Bedrock	`bedrock/llama-3-8b-instruct`

Bare slugs work in all of these cases. Use them when you don’t care which provider serves the request. Use the provider/ prefix when you specifically want to pin to one host (for ZDR compliance, contractual reasons, or latency in a specific region). The only universally required slug change is auto-routing:

Portkey	Concentrate AI
`model: "auto"` inside a Config strategy	`model: "auto"` with explicit `routing.model.sort` (`cost` / `latency` / `performance`)

For the authoritative list of supported provider/model-id pairs, call GET /v1/models or browse the Model Fortress.

Step 6: Reconnect Observability

Concentrate’s dashboard covers the major surfaces you used in Portkey.

Portkey dashboard surface	Concentrate equivalent
Logs (full request/response)	Per-request logs at concentrate.ai
Traces (`trace_id`-grouped)	`previous_response_id` on the Responses API links related requests into a stateful tree
Metadata filters (`_user`, `_environment`, custom keys)	Per-user / team / organization keys. Analytics roll up by whichever key, team, or org the request was made with
Feedback API (thumbs up/down)	No equivalent today
Analytics (cost, latency, tokens, cache hit rate)	Org / team / developer / key spend and latency rollups
Rate Limits / Usage Limits Policies	Per-key rate and spend limits in the dashboard
Audit Logs	Available in the dashboard
Log Exports	Available via the dashboard

Exporting your Portkey history

Portkey’s dashboards do not import into Concentrate, and the reverse is also true. Historical request logs stay where they were created. If your migration is driven by compliance or audit requirements, Portkey exposes Log Exports via their Admin API; pull your history out before deprovisioning your Portkey workspace.

Step 7: Migrate Admin and Governance Surface

Portkey’s Admin API (Workspaces, Virtual Keys, Configs, Prompts, Policies, SCIM mappings) maps roughly to Concentrate’s organization/team/key hierarchy plus dashboard-managed settings rather than a separate API surface.

Portkey Admin concept	Concentrate equivalent
Workspace	Team (within an organization)
Virtual Key / Provider Integration	Not exposed. Managed credentials
Config	Body params per request
API Key	API Key (per developer, per team, or per org)
Rate Limits / Usage Limits Policy	Per-key rate and spend limits
User / User Invite / Workspace Member	Organization and team members in the dashboard
SCIM Workspace Mappings	Available for enterprise plans. Contact support@concentrate.ai
Audit Logs	Available in the dashboard
Prompts / Partials / Labels / Collections	No equivalent today
Guardrails	Guardrails & Redaction
Secret References	Not applicable. Concentrate owns provider credentials

Step 8 (Optional): Adopt the Responses API

For new code, we recommend the native Responses API. It supports streaming, tool calling, structured output, multi-modal input, and web search through a single normalized shape across every provider, and previous_response_id replaces Portkey’s trace_id-grouped sessions with a server-managed conversation tree.

curl https://api.concentrate.ai/v1/responses \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $CONCENTRATE_API_KEY" \
  -d '{
    "model": "anthropic/claude-opus-4-6",
    "input": "What is the capital of France?"
  }'

import os
import requests

response = requests.post(
    "https://api.concentrate.ai/v1/responses",
    headers={
        "Authorization": f"Bearer {os.environ['CONCENTRATE_API_KEY']}",
        "Content-Type": "application/json",
    },
    json={
        "model": "anthropic/claude-opus-4-6",
        "input": "What is the capital of France?",
    },
)
print(response.json())

const response = await fetch("https://api.concentrate.ai/v1/responses", {
  method: "POST",
  headers: {
    "Authorization": `Bearer ${process.env.CONCENTRATE_API_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    model: "anthropic/claude-opus-4-6",
    input: "What is the capital of France?",
  }),
});
console.log(await response.json());

Why migrate to Concentrate

How it works

Portkey offers two integration shapes. The migration path differs for each:

BYOK mode. Point the OpenAI SDK at https://api.portkey.ai/v1, set x-portkey-api-key, and pass your own provider key via Authorization (with x-portkey-provider: openai) or a stored Virtual Key (x-portkey-virtual-key, or x-portkey-provider: @provider-slug). Migrating here gets you off provider keys entirely.
Config-driven mode. Use x-portkey-config to reference a saved or inline JSON config covering caching, fallbacks, load balancing, retries, and timeouts. Decompose the Config: most settings become Concentrate body params or are handled automatically. See Step 4.

Team-scale spend management

Concentrate organizes billing around an organization → team → developer → key hierarchy. Set budgets at any level and roll spend up into a single dashboard. Per-team budgets and per-developer attribution come from the key itself, so there’s no per-request metadata tagging to maintain.

Feature-aware resiliency

Beyond ordered model and provider fallbacks (routing.model.fallbacks, routing.provider.fallbacks), Concentrate’s routing layer ships:

Uptime gate. Providers whose per-feature success rate drops below 90% are skipped.
Feature degradation. If no provider supports the full requested feature set (e.g. json_schema), the request is downgraded to json_object or text instead of failing.
Cache-affinity routing. When multiple providers can serve a request, the one where your actor already has cached tokens is preferred.

All on by default. No Config object to author or version.

Strategy-driven auto routing

model: "auto" accepts an explicit optimization target via routing.model.sort: cost, latency, or performance (default). See Auto Routing.

Native Responses and Messages APIs

Alongside OpenAI Chat Completions, Concentrate exposes a first-class Responses API and an Anthropic-compatible Messages API.

Managed provider credentials by default

Concentrate manages provider credentials by default. There are no Virtual Keys, Provider Integrations, or Model Catalog you must configure — point at a model and Concentrate owns the upstream credentials. If you’d rather keep using your own provider accounts, store those keys once in the dashboard with free BYOK and routing uses them automatically.

Troubleshooting

Model not found

Bare slugs (gpt-4o, claude-haiku-4-5) and provider/model-id slugs both work. If you’re using a provider/ prefix and getting a miss, double-check the prefix is a provider (e.g. bedrock, azure, ai-studio) and not just the author (e.g. meta, google). Call GET /v1/models for the authoritative list.

Invalid API key error

Concentrate keys start with sk-cn-v1-. If you are still sending a Portkey x-portkey-api-key value (or an upstream provider key from BYOK mode) as the Authorization bearer, you will see a 401. Verify the value in your dashboard and confirm there are no extra spaces or quotes.

Requests succeed but nothing shows up in the Concentrate dashboard

Confirm the base URL is https://api.concentrate.ai/v1, not api.portkey.ai. If the SDK is still pointed at Portkey it is logging against your Portkey workspace, not Concentrate.

My metadata filters stopped working

Concentrate does not accept arbitrary per-request metadata. Reserved Portkey keys like _user, _environment, and _organisation map to the key/team/org hierarchy. Issue a separate key per user or environment and analytics roll up automatically. Custom metadata keys have no equivalent today.

My Config-based routing stopped applying

x-portkey-config is a no-op on Concentrate. Express fallbacks via routing.model.fallbacks / routing.provider.fallbacks body params, or use model: "auto" with a routing strategy. Conditional routing by metadata has no direct equivalent. Handle in application code.

Cache hit rate dropped after migrating

Concentrate uses provider-native prompt caching, currently supported on Anthropic and AWS Bedrock. There is no semantic cache and no gateway-stored cache, so Portkey’s simple and semantic cache modes do not carry over. Caches are seeded per API key by default; pass prompt_cache_key in the request body if you want to set the seed explicitly (the analog of x-portkey-cache-namespace).

Connection errors

Confirm the base URL is https://api.concentrate.ai/v1 (no /api segment, no per-provider subdomain). Test the connection manually:

curl https://api.concentrate.ai/v1/responses/health

Next Steps

API Reference

Explore the full API capabilities

Available Models

Browse all supported models

Auto Routing

Optimize model selection automatically

Get Support

Contact our support team

Feedback

If you hit anything that didn’t translate cleanly (especially around Configs, Prompt templates, conditional routing, custom hosts, or per-request metadata), email support@concentrate.ai. The capability gaps called out above are tracked, and migration friction reports directly shape what we ship next.

Getting started

Migrations

Alerts

Prerequisites

Quick Start for Claude Code users

Step 1: Update Your Environment Variables

Step 2: Update Your Client

Step 3: Remove `x-portkey-*` Headers

Step 4: Decompose Your Portkey Config

Step 5: Update Model Identifiers

Step 6: Reconnect Observability

Exporting your Portkey history

Step 7: Migrate Admin and Governance Surface

Step 8 (Optional): Adopt the Responses API

Why migrate to Concentrate

Troubleshooting

Next Steps

API Reference

Available Models

Auto Routing

Get Support

Feedback

​Prerequisites

​Quick Start for Claude Code users

​Step 1: Update Your Environment Variables

​Step 2: Update Your Client

​Step 3: Remove x-portkey-* Headers

​Step 4: Decompose Your Portkey Config

​Step 5: Update Model Identifiers

​Step 6: Reconnect Observability

​Exporting your Portkey history

​Step 7: Migrate Admin and Governance Surface

​Step 8 (Optional): Adopt the Responses API

​Why migrate to Concentrate

​Troubleshooting

​Next Steps

API Reference

Available Models

Auto Routing

Get Support

​Feedback

Prerequisites

Quick Start for Claude Code users

Step 1: Update Your Environment Variables

Step 2: Update Your Client

Step 3: Remove `x-portkey-*` Headers

Step 4: Decompose Your Portkey Config

Step 5: Update Model Identifiers

Step 6: Reconnect Observability

Exporting your Portkey history

Step 7: Migrate Admin and Governance Surface

Step 8 (Optional): Adopt the Responses API

Why migrate to Concentrate

Troubleshooting

Next Steps

Feedback