M

Meta: Llama 3.1 70B Instruct

Meta
128K
Context
$0.4000
Input /1M
$0.4000
Output /1M
N/A
Max Output
Run API Call Open in Chat

Meta: Llama 3.1 70B Instruct

Description

Meta's latest language model offering comes in various sizes. This 70B parameter instruct-tuned variant targets high-quality dialogue applications. The model has demonstrated strong performance compared to leading closed-source models in human evaluations.

This instruct-tuned variant represents Meta's latest generation of large language models, optimized for dialogue and conversation tasks.

Technical Specifications

Specification Value
Context Window 128,000 tokens
Context Length 131,072 tokens
Input Modalities Text
Output Modalities Text
Instruct Type llama3
Default Stop Sequences <|eot_id|>, <|end_of_text|>

Pricing

Type Price
Input Tokens $0.40 per 1M tokens
Output Tokens $0.40 per 1M tokens

Supported Parameters

The following parameters are supported for inference:

  • max_tokens - Maximum number of tokens to generate
  • temperature - Controls randomness in generation (0.0 to 2.0)
  • top_p - Nucleus sampling parameter
  • stop - Custom stop sequences
  • frequency_penalty - Reduces repetition of frequent tokens
  • presence_penalty - Reduces repetition of any tokens already present
  • seed - Random seed for reproducible outputs
  • logit_bias - Adjust likelihood of specific tokens
  • top_k - Top-k sampling parameter
  • min_p - Minimum probability threshold
  • repetition_penalty - Alternative repetition control

API Usage Example

curl -X POST https://api.langmart.ai/v1/chat/completions \
  -H "Authorization: Bearer $LANGMART_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "meta-llama/llama-3.1-70b-instruct",
    "messages": [
      {"role": "user", "content": "Hello, how are you?"}
    ],
    "temperature": 0.7,
    "max_tokens": 1024
  }'

Providers

Primary Provider: Hyperbolic

Feature Value
Provider ID hyperbolic/fp8
Quantization FP8
Tool Choice Support literal_none, literal_auto, literal_required, type_function
Endpoints Chat Completions, Completions

Model Identification

Property Value
Model Name Meta: Llama 3.1 70B Instruct
Model ID meta-llama/llama-3.1-70b-instruct
Author/Creator Meta (meta-llama)
Release Date July 23, 2024
Model Size 70 Billion Parameters

Model Capabilities

  • High-quality dialogue and conversation
  • Instruction following
  • Tool use / Function calling
  • Long context understanding (up to 131K tokens)
  • Multi-turn conversations

Usage Statistics

The model processes substantial daily traffic on LangMart, with recent analytics showing:

  • Millions of requests processed daily
  • Billions of tokens processed (December 2025 data indicates ~2M+ daily requests)
  • meta-llama/llama-3.1-8b-instruct - Smaller 8B parameter variant
  • meta-llama/llama-3.1-405b-instruct - Larger 405B parameter variant

License and Compliance

Usage of this model is subject to Meta's Acceptable Use Policy. Users must comply with Meta's licensing terms and acceptable use guidelines.

Notes

  • The model uses the llama3 instruct format for prompting
  • FP8 quantization is used by the primary provider (Hyperbolic) for efficient inference
  • Supports streaming responses
  • Function calling / tool use is supported with various tool choice options

Data sourced from LangMart.ai on December 23, 2025