⚡️ FlashAttention-4: up to 1.3× faster than cuDNN on NVIDIA Blackwell →

Introducing Together AI's new look →

🔎 ATLAS: runtime-learning accelerators delivering up to 4x faster LLM inference →

⚡ Together GPU Clusters: self-service NVIDIA GPUs, now generally available →

📦 Batch Inference API: Process billions of tokens at 50% lower cost for most models →

🪛 Fine-Tuning Platform Upgrades: Larger Models, Longer Contexts →

Models / Meta
LLM
Chat

Llama 3 70B Instruct Reference

Auto-regressive LLM with optimized transformers, SFT, and RLHF for alignment with helpfulness and safety preferences.

This model is not available on Together’s Serverless API.

Pick a supported alternative from the Model Library.

Related models
new
Chat
GLM-5.1

  • Input/1M

    $1.40

  • Output/1M

    $4.40
  • Parameters
    754B
Function Calling
JSON Mode
Reasoning
Chat
New
Chat
MiniMax M2.7

  • Cached/1M

    $0.06

  • Input/1M

    $0.30

  • Output/1M

    $1.20
Function Calling
JSON Mode
Prompt Caching
Reasoning
Code
Chat
new
Chat
Kimi K2.6

  • Cached/1M

    $0.20

  • Input/1M

    $1.20

  • Output/1M

    $4.50
Function Calling
JSON Mode
Reasoning
Vision
Chat
Code
LLM
Model specifications
Model data
  • Model provider
    Meta
  • Type
    LLM
    Chat
  • Main use cases
    Chat
  • Features
    JSON Mode
  • Parameters
    70B
  • Context length
    8K