from transformers import AutoModelForCausalLM, AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained("nvidia/embed-qa-4")
model = AutoModelForCausalLM.from_pretrained("nvidia/embed-qa-4")
import { pipeline } from '@huggingface/transformers';
const generator = await pipeline('text-generation', 'nvidia/embed-qa-4');
const output = await generator('Hello, I am a language model', { max_new_tokens: 50 });
// Using Hugging Face Inference API
import "github.com/hupe1980/go-huggingface"
client := huggingface.NewInferenceClient("HF_API_TOKEN")
ic, err := client.TextGeneration(context.Background(), &huggingface.TextGenerationRequest{
Model: "nvidia/embed-qa-4",
Inputs: "Hello, I am a language model",
})
// Using HF candle crate
use candle_transformers::models::llama;
let api = Api::new()?;
let repo = api.model("nvidia/embed-qa-4".to_string());
// Load model using llama.cpp
#include "llama.h"
llama_model_params model_params = llama_model_default_params();
llama_model* model = llama_model_load_from_file("nvidia/embed-qa-4.gguf", model_params);
<?php
// Call HF Inference API for text generation
$ch = curl_init("https://api-inference.huggingface.co/models/nvidia/embed-qa-4");
curl_setopt($ch, CURLOPT_HTTPHEADER, ["Authorization: Bearer HF_API_TOKEN"]);
curl_setopt($ch, CURLOPT_POSTFIELDS, json_encode(["inputs" => "Hello, I am a language model"]));
from transformers import AutoModelForCausalLM, AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained("nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4")
model = AutoModelForCausalLM.from_pretrained("nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4")
import { pipeline } from '@huggingface/transformers';
const generator = await pipeline('text-generation', 'nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4');
const output = await generator('Hello, I am a language model', { max_new_tokens: 50 });
// Using Hugging Face Inference API
import "github.com/hupe1980/go-huggingface"
client := huggingface.NewInferenceClient("HF_API_TOKEN")
ic, err := client.TextGeneration(context.Background(), &huggingface.TextGenerationRequest{
Model: "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4",
Inputs: "Hello, I am a language model",
})
// Using HF candle crate
use candle_transformers::models::llama;
let api = Api::new()?;
let repo = api.model("nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4".to_string());
// Load model using llama.cpp
#include "llama.h"
llama_model_params model_params = llama_model_default_params();
llama_model* model = llama_model_load_from_file("nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4.gguf", model_params);
<?php
// Call HF Inference API for text generation
$ch = curl_init("https://api-inference.huggingface.co/models/nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4");
curl_setopt($ch, CURLOPT_HTTPHEADER, ["Authorization: Bearer HF_API_TOKEN"]);
curl_setopt($ch, CURLOPT_POSTFIELDS, json_encode(["inputs" => "Hello, I am a language model"]));
