Llama 2 70B
Meta's previous generation flagship open-source model
Llama 2 70B
Meta • July 2023
Training Data
Up to early 2023
Llama 2 70B
July 2023
Parameters
70 billion
Training Method
Reinforcement Learning
Context Window
4,096 tokens
Knowledge Cutoff
September 2022
Key Features
Open Source • Commercial Use • Safety Focused
Capabilities
Reasoning: Good
Safety: Very Good
General: Good
What's New in This Version
Improved safety and helpfulness over Llama 1
Meta's previous generation flagship open-source model
What's New in This Version
Improved safety and helpfulness over Llama 1
Technical Specifications
Key Features
Capabilities
Other Meta Models
Explore more models from Meta
Muse Spark 1.1
Meta Superintelligence Labs' strongest model for agentic and coding work — the first Muse model developers can build on, via the new Meta Model API
Muse Spark
Meta Superintelligence Labs' first model — a natively multimodal reasoning model with contemplating mode and multi-agent orchestration
Llama 4 Behemoth
Meta's flagship multimodal model (RESEARCH PREVIEW ONLY - weights not publicly released)
