OpenRouter

12 min read

Updated 2026-04-09

Migrating from OpenRouter to AI Stats Gateway

Move from OpenRouter to AI Stats Gateway with a safe rollout plan, code-level before/after examples, model-ID checks, and validation commands for non-streaming and streaming traffic.

Prerequisites

Access to your current OpenRouter integration code and deployment config.
An AI Stats API key stored as `AI_STATS_API_KEY` in each target environment.
A list of production model IDs currently used by your application.

1) Inventory current OpenRouter usage

Start by identifying every place OpenRouter is referenced: endpoint URLs, keys, model IDs, and provider-specific headers.

Keeping the migration boundary small lowers risk. Usually, updating one gateway client module is enough.

Checklist

Find `openrouter.ai` endpoint references.
Find `OPENROUTER_API_KEY` usage in code, CI, and hosting env vars.
Find OpenRouter-only headers such as `HTTP-Referer` and `X-Title`.
Document currently used model IDs and any fallback chain logic.

Screenshot checkpoints

Repository search results

Capture your code search showing all OpenRouter references before changes.

Suggested asset path: /migrate/openrouter/01-openrouter-search.png

2) Switch base URL and credentials

Replace the OpenRouter base URL with AI Stats and move authentication to `AI_STATS_API_KEY`.

Keep request payload shape unchanged first. Do behavior parity before optimization.

Before (OpenRouter + OpenAI SDK)

import OpenAI from "openai";

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

const response = await client.chat.completions.create({
  model: "openai/gpt-4.1-mini",
  messages: [{ role: "user", content: "Summarize our migration plan." }],
});

After (AI Stats Gateway + OpenAI SDK)

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.AI_STATS_API_KEY,
  baseURL: "https://api.phaseo.app/v1",
});

const response = await client.chat.completions.create({
  model: "openai/gpt-4.1-mini",
  messages: [{ role: "user", content: "Summarize our migration plan." }],
});

Environment variable rename

# Before
OPENROUTER_API_KEY=...

# After
AI_STATS_API_KEY=...

Screenshot checkpoints

Environment variable update

Capture your hosting dashboard after adding `AI_STATS_API_KEY` and removing `OPENROUTER_API_KEY`.

Suggested asset path: /migrate/openrouter/02-env-vars.png

3) Validate model IDs and remove OpenRouter-specific behavior

Do not assume every old model alias is valid. Query `/v1/models` and verify each production model ID.

If your app consumed OpenRouter-specific response fields, adapt at one compatibility layer instead of changing every caller.

Checklist

Keep `Authorization: Bearer` format unchanged.
Retain attribution headers only if you still need them for analytics.
Test both non-streaming and streaming chat paths before rollout.

Check model list from AI Stats

curl -s "https://api.phaseo.app/v1/models" \
  -H "Authorization: Bearer $AI_STATS_API_KEY" | jq '.data[0:10] | map(.id)'

Optional model alias compatibility map

const MODEL_ALIASES: Record<string, string> = {
  "openai/gpt-4.1-mini": "openai/gpt-4.1-mini",
  "anthropic/claude-3.5-sonnet": "anthropic/claude-3.5-sonnet",
};

export function resolveModelId(input: string): string {
  return MODEL_ALIASES[input] ?? input;
}

4) Roll out safely

Use a staged rollout: dev first, then a small production slice, then full traffic once metrics are stable.

Track latency, error rate, and token/cost drift. Roll back by switching only endpoint+key config if needed.

Checklist

Start with internal traffic only.
Move to 5-10% production traffic and compare quality/cost metrics.
Promote to 100% after parity is confirmed.

Screenshot checkpoints

Production canary metrics

Capture dashboard or logs showing error rate and latency during canary rollout.

Suggested asset path: /migrate/openrouter/03-canary-metrics.png

Validation Steps

curl -s "https://api.phaseo.app/v1/health"
curl -s "https://api.phaseo.app/v1/models" -H "Authorization: Bearer $AI_STATS_API_KEY"
curl -s "https://api.phaseo.app/v1/chat/completions" -H "Content-Type: application/json" -H "Authorization: Bearer $AI_STATS_API_KEY" -d '{"model":"openai/gpt-4.1-mini","messages":[{"role":"user","content":"Say hello"}]}'
Run one streaming request through your app-level integration test.
Run one negative test (invalid key or invalid model) to verify failure handling.

FAQ

Do I need to rewrite prompts or message payloads?

No. Most teams keep payloads the same and only switch base URL, key source, and optional model alias mapping.

Can I keep provider-prefixed model IDs?

Often yes, but verify against `/v1/models`. If your existing aliases differ, normalize them in one boundary function.

Need a custom migration diff?

Use the interactive assistant for before/after snippets tailored to your current SDK and language.

Open Migration Assistant

OpenRouter

12 min read

Updated 2026-04-09

Migrating from OpenRouter to AI Stats Gateway

Move from OpenRouter to AI Stats Gateway with a safe rollout plan, code-level before/after examples, model-ID checks, and validation commands for non-streaming and streaming traffic.

Prerequisites

Access to your current OpenRouter integration code and deployment config.
An AI Stats API key stored as `AI_STATS_API_KEY` in each target environment.
A list of production model IDs currently used by your application.

1) Inventory current OpenRouter usage

Start by identifying every place OpenRouter is referenced: endpoint URLs, keys, model IDs, and provider-specific headers.

Keeping the migration boundary small lowers risk. Usually, updating one gateway client module is enough.

Checklist

Find `openrouter.ai` endpoint references.
Find `OPENROUTER_API_KEY` usage in code, CI, and hosting env vars.
Find OpenRouter-only headers such as `HTTP-Referer` and `X-Title`.
Document currently used model IDs and any fallback chain logic.

Screenshot checkpoints

Repository search results

Capture your code search showing all OpenRouter references before changes.

Suggested asset path: /migrate/openrouter/01-openrouter-search.png

2) Switch base URL and credentials

Replace the OpenRouter base URL with AI Stats and move authentication to `AI_STATS_API_KEY`.

Keep request payload shape unchanged first. Do behavior parity before optimization.

Before (OpenRouter + OpenAI SDK)

import OpenAI from "openai";

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

const response = await client.chat.completions.create({
  model: "openai/gpt-4.1-mini",
  messages: [{ role: "user", content: "Summarize our migration plan." }],
});

After (AI Stats Gateway + OpenAI SDK)

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.AI_STATS_API_KEY,
  baseURL: "https://api.phaseo.app/v1",
});

const response = await client.chat.completions.create({
  model: "openai/gpt-4.1-mini",
  messages: [{ role: "user", content: "Summarize our migration plan." }],
});

Environment variable rename

# Before
OPENROUTER_API_KEY=...

# After
AI_STATS_API_KEY=...

Screenshot checkpoints

Environment variable update

Capture your hosting dashboard after adding `AI_STATS_API_KEY` and removing `OPENROUTER_API_KEY`.

Suggested asset path: /migrate/openrouter/02-env-vars.png

3) Validate model IDs and remove OpenRouter-specific behavior

Do not assume every old model alias is valid. Query `/v1/models` and verify each production model ID.

If your app consumed OpenRouter-specific response fields, adapt at one compatibility layer instead of changing every caller.

Checklist

Keep `Authorization: Bearer` format unchanged.
Retain attribution headers only if you still need them for analytics.
Test both non-streaming and streaming chat paths before rollout.

Check model list from AI Stats

curl -s "https://api.phaseo.app/v1/models" \
  -H "Authorization: Bearer $AI_STATS_API_KEY" | jq '.data[0:10] | map(.id)'

Optional model alias compatibility map

const MODEL_ALIASES: Record<string, string> = {
  "openai/gpt-4.1-mini": "openai/gpt-4.1-mini",
  "anthropic/claude-3.5-sonnet": "anthropic/claude-3.5-sonnet",
};

export function resolveModelId(input: string): string {
  return MODEL_ALIASES[input] ?? input;
}

4) Roll out safely

Use a staged rollout: dev first, then a small production slice, then full traffic once metrics are stable.

Track latency, error rate, and token/cost drift. Roll back by switching only endpoint+key config if needed.

Checklist

Start with internal traffic only.
Move to 5-10% production traffic and compare quality/cost metrics.
Promote to 100% after parity is confirmed.

Screenshot checkpoints

Production canary metrics

Capture dashboard or logs showing error rate and latency during canary rollout.

Suggested asset path: /migrate/openrouter/03-canary-metrics.png

Validation Steps

curl -s "https://api.phaseo.app/v1/health"
curl -s "https://api.phaseo.app/v1/models" -H "Authorization: Bearer $AI_STATS_API_KEY"
curl -s "https://api.phaseo.app/v1/chat/completions" -H "Content-Type: application/json" -H "Authorization: Bearer $AI_STATS_API_KEY" -d '{"model":"openai/gpt-4.1-mini","messages":[{"role":"user","content":"Say hello"}]}'
Run one streaming request through your app-level integration test.
Run one negative test (invalid key or invalid model) to verify failure handling.

FAQ

Do I need to rewrite prompts or message payloads?

No. Most teams keep payloads the same and only switch base URL, key source, and optional model alias mapping.

Can I keep provider-prefixed model IDs?

Often yes, but verify against `/v1/models`. If your existing aliases differ, normalize them in one boundary function.

Need a custom migration diff?

Use the interactive assistant for before/after snippets tailored to your current SDK and language.

Open Migration Assistant