NVIDIA NeMo

# NVIDIA uses OpenAI client to interact with their API
from openai import OpenAI  
from honeyhive import HoneyHiveTracer

# place the code below at the beginning of your application execution
HoneyHiveTracer.init(
    api_key="MY_HONEYHIVE_API_KEY", # paste your API key here
    project="MY_HONEYHIVE_PROJECT_NAME", # paste your project name here
)

client = OpenAI(
    base_url="https://integrate.api.nvidia.com/v1",
    api_key="MY_NVIDIA_API_KEY",
)

completion = client.chat.completions.create(
    model="nvidia/mistral-nemo-minitron-8b-8k-instruct",
    messages=[
        {
            "role": "user",
            "content": "Write a limerick about the wonders of GPU computing.",
        }
    ],
    stream=True,
)

for chunk in completion:
    if chunk.choices[0].delta.content is not None:
        print(chunk.choices[0].delta.content, end="")

import OpenAI from 'openai';
import { HoneyHiveTracer } from 'honeyhive';

const tracer = await HoneyHiveTracer.init({
    apiKey: 'MY_HONEYHIVE_API_KEY',
    project: 'MY_HONEYHIVE_PROJECT_NAME',
    sessionName: 'test',
});

const openai = new OpenAI({
    apiKey: "MY_NVIDIA_API_KEY",
    baseURL: 'https://integrate.api.nvidia.com/v1',
})

async function main(input: string): Promise<string> {
    const completion = await openai.chat.completions.create({
        model: "nvidia/mistral-nemo-minitron-8b-8k-instruct",
        messages: [{ role: "user" as const, content: input }],
        temperature: 0.2,
        top_p: 0.7,
        max_tokens: 1024,
        stream: true
    });

    let ret = '';
    let content: string;
    
    for await (const chunk of completion) {
        content = chunk.choices[0]?.delta?.content ?? '\n';
        process.stdout.write(content);
        ret += content;
    }
    return ret;
}

const tracedMain = tracer.traceFunction()(main);
await tracedMain("Write a limerick about the wonders of GPU computing.");