Skip to main content

​
About the Provider

Meta is a leading global technology company focusing on social media, connectivity, and artificial intelligence research. Meta develops advanced AI models, such as the LLaMA family, to empower developers and enterprises with scalable language understanding and generation capabilities. Its open-weight AI initiatives aim to foster innovation and broader community access to powerful AI tools.

​
Model Quickstart

This section helps you quickly get started with the meta-llama/Llama-3.3-70B-Instruct model on the Qubrid AI inferencing platform. To use this model, you need:
  • A valid Qubrid API key
  • Access to the Qubrid inference API
  • Basic knowledge of making API requests in your preferred language
Once authenticated with your API key, you can send inference requests to the meta-llama/Llama-3.3-70B-Instruct model and receive responses based on your input prompts. Below are example placeholders showing how the model can be accessed using different programming environments.
You can choose the one that best fits your workflow. 
import requests
import json
from pprint import pprint

url = "https://platform.qubrid.com/api/v1/qubridai/chat/completions"

headers = {
  "Authorization": "Bearer <QUBRID_API_KEY>",
  "Content-Type": "application/json"
}

data = {
  "model": "meta-llama/Llama-3.3-70B-Instruct",
  "messages": [
      {
          "role": "user",
          "content": "Explain quantum computing to a 5 year old."
      }
  ],
  "temperature": 0.7,
  "max_tokens": 4096,
  "stream": False,
  "top_p": 0.8
}

response = requests.post(
  url,
  headers=headers,
  json=data,
)

content_type = response.headers.get("Content-Type", "")

if "application/json" in content_type:
  pprint(response.json())
else:
  for line in response.iter_lines(decode_unicode=True):
      if not line:
          continue

      if line.startswith("data:"):
          payload = line.replace("data:", "").strip()

          if payload == "[DONE]":
              break

          try:
              chunk = json.loads(payload)
              pprint(chunk)
          except json.JSONDecodeError:
              print("Raw chunk:", payload)

​
Model Overview

Llama 3.3 70B Instruct is a 70B-parameter open-weight large language model from Meta, optimized for instruction following, complex reasoning, and multi-turn conversations.It is well suited for enterprise use cases such as advanced chat assistants, code reasoning, and long-document analysis with large context windows.

​
Model at a Glance

FeatureDetails
Model IDLlama-3.3-70B-Instruct
ArchitectureTransformer with Grouped-Query Attention(GQA)
Model Size70B parameters
Parameters4
Training DataPublicly available web data (multilingual)
Context Length128K Token

​
Supported languages

  • English
  • German
  • French
  • Italian
  • Portuguese
  • Hindi
  • Spanish
  • Thai

​
When to use?

Use Llama 3.3 70B Instruct if you need:
  • Enterprise chat assistants
  • Advanced code generation and review
  • Long-document question answering
  • Summarization at scale
  • Retrieval-Augmented Generation (RAG)
  • AI agents and workflow automation

​
Inference Parameters

Parameter NameTypeDefaultDescription
StreamingbooleantrueEnable streaming responses for real-time output.
Temperaturenumber0.7Controls randomness. Higher values mean more creative but less predictable output.
Max Tokensnumber4096Defines the maximum number of tokens the model is allowed to generate.
Top Pnumber0.9Nucleus sampling that limits token selection to a subset of top probability mass.

​
Key Features

  1. High-quality reasoning and instruction adherence
  2. Strong performance on code and analytical tasks
  3. Large context window for long-document processing
  4. Open-weight model suitable for private and on-prem deployments
  5. Production-ready for enterprise workloads

​
Limitations

  1. Smaller context window compared to largest models
  2. Can struggle with highly complex, multi-step reasoning

​
Summary

Meta Llama 3.3 70B Instruct is a 70B-parameter, instruction-tuned large language model designed for high-quality reasoning and multi-turn conversational tasks. It is well suited for enterprise workloads such as advanced chat assistants, code generation, summarization, and long-document question answering. The model supports a large context window, enabling effective processing of lengthy inputs and retrieval-augmented generation workflows.