Lepton

Lepton AI provides efficient inference for open-source language models with optimized deployment and scaling. Braintrust integrates seamlessly with Lepton through direct API access, wrapper functions for automatic tracing, and proxy support.

Setup

To use Lepton models, configure your Lepton API key in Braintrust.

Get a Lepton API key from Lepton AI Dashboard
Add the Lepton API key to your organization's AI providers
Set the Lepton API key and your Braintrust API key as environment variables

.env

LEPTON_API_KEY=<your-lepton-api-key>
BRAINTRUST_API_KEY=<your-braintrust-api-key>
 
# If you are self-hosting Braintrust, set the URL of your hosted dataplane
# BRAINTRUST_API_URL=<your-braintrust-api-url>

API keys are encrypted using 256-bit AES-GCM encryption and are not stored or logged by Braintrust.

Use Lepton with Braintrust AI proxy

The Braintrust AI Proxy allows you to access Lepton models through a unified OpenAI-compatible interface.

Install the braintrust and openai packages.

pnpm add braintrust openai

Then, initialize the client and make a request to a Lepton model via the Braintrust AI Proxy.

lepton_proxy.ts

import { OpenAI } from "openai";
 
const client = new OpenAI({
  baseURL: "https://api.braintrust.dev/v1/proxy",
  apiKey: process.env.BRAINTRUST_API_KEY,
});
 
const response = await client.chat.completions.create({
  model: "openai/gpt-oss-120b",
  messages: [{ role: "user", content: "Hello, world!" }],
});

Trace logs with Lepton

Trace your Lepton LLM calls for observability and monitoring.

When using the Braintrust AI Proxy, API calls are automatically logged to the specified project.

lepton_trace.ts

import { OpenAI } from "openai";
import { initLogger } from "braintrust";
 
initLogger({
  projectName: "My Project",
  apiKey: process.env.BRAINTRUST_API_KEY,
});
 
const client = new OpenAI({
  baseURL: "https://api.braintrust.dev/v1/proxy",
  apiKey: process.env.BRAINTRUST_API_KEY,
});
 
// All API calls are automatically logged
const result = await client.chat.completions.create({
  model: "openai/gpt-oss-120b",
  messages: [{ role: "user", content: "What is machine learning?" }],
});

The Braintrust AI Proxy is not required to trace Lepton API calls. For more control, learn how to customize traces.

Evaluate with Lepton

Evaluations distill the non-deterministic outputs of Lepton models into an effective feedback loop that enables you to ship more reliable, higher quality products. Braintrust Eval is a simple function composed of a dataset of user inputs, a task, and a set of scorers. To learn more about evaluations, see the Experiments guide.

lepton_eval.ts

import { Eval } from "braintrust";
import { OpenAI } from "openai";
 
const client = new OpenAI({
  baseURL: "https://api.braintrust.dev/v1/proxy",
  apiKey: process.env.BRAINTRUST_API_KEY,
});
 
Eval("Lepton Evaluation", {
  data: () => [
    { input: "What is 2+2?", expected: "4" },
    { input: "What is the capital of France?", expected: "Paris" },
  ],
  task: async (input) => {
    const response = await client.chat.completions.create({
      model: "llama3-3-70b",
      messages: [{ role: "user", content: input }],
    });
    return response.choices[0].message.content;
  },
  scores: [
    {
      name: "accuracy",
      scorer: (args) => (args.output === args.expected ? 1 : 0),
    },
  ],
});

To learn more about tool use, multimodal support, attachments, and masking sensitive data with Lepton, visit the customize traces guide.

Models and capabilities

Model	Max input	Max output	Input $/1M	Output $/1M
llama3-3-70b	﹣	﹣	$0.80	$0.80
llama3-2-3b	﹣	﹣	$0.03	$0.03
llama3-2-1b	﹣	﹣	$0.01	$0.01
llama3-1-70b	﹣	﹣	$0.80	$0.80
llama3-1-8b	﹣	﹣	$0.07	$0.07
llama3-70b	﹣	﹣	$0.80	$0.80
llama3-8b	﹣	﹣	$0.07	$0.07
mistral-7b	﹣	﹣	$0.07	$0.07
mixtral-8x7b	﹣	﹣	$0.50	$0.50
wizardlm-2-7b	﹣	﹣	$0.07	$0.07
wizardlm-2-8x22b	﹣	﹣	$1.00	$1.00
nous-hermes-llama2-13b	﹣	﹣	$0.18	$0.18
dolphin-mixtral-8x7b	﹣	﹣	$0.50	$0.50