🚪 AI Gateway:
- Unified API Signature: If you've used OpenAI, you already know how to use Portkey with any other provider.
- Interoperability: Write once, run with any provider. Switch between any model from any provider seamlessly.
- Automated Fallbacks & Retries: Ensure your application remains functional even if a primary service fails.
- Load Balancing & A/B Testing: Efficiently distribute incoming requests among multiple models and run A/B tests at scale.
- Semantic Caching: Reduce costs and latency by intelligently caching results.
🔬 Observability:
- Logging: Keep track of all requests for monitoring and debugging.
- Requests Tracing: Understand the journey of each request for optimization.
- Custom Tags: Segment and categorize requests for better insights.
$ npm install portkey-ai
$ export PORTKEY_API_KEY="PORTKEY_API_KEY"import { Portkey } from "portkey-ai";
const portkey = new Portkey({
mode: "single",
llms: [{ provider: "openai", virtual_key: "open-ai-xxx" }]
});
async function main() {
const chatCompletion = await portkey.chat.completions.create({
messages: [{ role: 'user', content: 'Say this is a test' }],
model: 'gpt-4'
});
console.log(chatCompletion.choices);
};
main();Portkey fully adheres to the OpenAI SDK signature. This means that you can instantly switch to Portkey and start using Portkey's advanced production features right out of the box.
There are 4️ Steps to Integrate Portkey
- Setting your Portkey API key and your virtual key for AI providers.
- Constructing your LLM with Portkey features, provider features (and prompt!).
- Constructing the Portkey client and setting usage mode.
- Making your request!
Let's dive in! If you are an advanced user and want to directly jump to various full-fledged examples, click here.
Portkey API Key: Log into Portkey here, then click on the profile icon on top left and “Copy API Key”.
export PORTKEY_API_KEY="PORTKEY_API_KEY"Virtual Keys: Navigate to the "Virtual Keys" page on Portkey and hit the "Add Key" button. Choose your AI provider and assign a unique name to your key. Your virtual key is ready!
Portkey Features: You can find a comprehensive list of Portkey features here. This includes settings for caching, retries, metadata, and more.
Provider Features:
Portkey is designed to be flexible. All the features you're familiar with from your LLM provider, like top_p, top_k, and temperature, can be used seamlessly. Check out the complete list of provider features here.
Setting the Prompt Input:
This param lets you override any prompt that is passed during the completion call - set a model-specific prompt here to optimise the model performance. You can set the input in two ways. For models like Claude and GPT3, use prompt = (str), and for models like GPT3.5 & GPT4, use messages = [array].
Here's how you can combine everything:
import { LLMOptions } from "portkey-ai";
// Portkey Config
const provider = "openai";
const virtual_key = "open-ai-xxx";
const trace_id = "portkey_sdk_test";
const cache_status = "semantic";
// Model Params
const model = "gpt-4";
const temperature = 1;
// Prompt
const messages = [{"role": "user", "content": "Who are you?"}];
const llm_a: LLMOptions = {
provider: provider,
virtual_key: virtual_key,
cache_status: cache_status,
trace_id: trace_id,
model: model,
temperature: temperature,
messages: messages
};Portkey client's config takes 3 params: api_key, mode, llms.
api_key: You can set your Portkey API key here or with$ EXPORTas done above.mode: There are 3 modes - Single, Fallback, Loadbalance.- Single - This is the standard mode. Use it if you do not want Fallback OR Loadbalance features.
- Fallback - Set this mode if you want to enable the Fallback feature.
- Loadbalance - Set this mode if you want to enable the Loadbalance feature.
llms: This is an array where we pass our LLMs constructed using the LLMOptions interface.
import { Portkey } from "portkey-ai";
const portkey = new Portkey({ mode: "single", llms: [llm_a] });The Portkey client can do ChatCompletions and Completions calls.
Since our LLM is GPT4, we will use ChatCompletions:
async function main() {
const response = await portkey.chatCompletions.create({
messages: [{ "role": "user", "content": "Who are you ?"}]
});
console.log(response.choices[0].message);
};
main();You have integrated Portkey's Node SDK in just 4 steps!
You can set all of these features while constructing your LLMOptions object.
| Feature | Config Key | Value(Type) | Required |
|---|---|---|---|
| API Key OR Virtual Key | api_key OR virtual_key |
string |
✅ Required |
| Provider Name | provider |
openai, cohere, anthropic, azure-openai |
✅ Required |
| Model Name | model |
The relevant model name from the provider. For example, gpt-3.5-turbo OR claude-2 |
❔ Optional |
| Weight (For Loadbalance) | weight |
integer |
❔ Optional |
| Cache Type | cache_status |
simple, semantic |
❔ Optional |
| Force Cache Refresh | cache_force_refresh |
True, False (Boolean) |
❔ Optional |
| Cache Age | cache_age |
integer (in seconds) |
❔ Optional |
| Trace ID | trace_id |
string |
❔ Optional |
| Retries | retry |
integer [0,5] |
❔ Optional |
| Metadata | metadata |
json object More info |
❔ Optional |
| All Model Params | As per the model/provider | This is params like top_p, temperature, etc |
❔ Optional |
| Provider | Support Status | Supported Endpoints | |
|---|---|---|---|
 |
OpenAI | ✅ Supported | /completion, /chatcompletion |
 |
Azure OpenAI | ✅ Supported | /completion, /chatcompletion |
 |
Anthropic | ✅ Supported | /complete |
 |
Cohere | ✅ Supported | generate |
