Meta Llama 3 is out now and available on Replicate. This language model family comes in 8B and 70B parameter sizes, with context windows of 8K tokens. The models beat most other open source models on industry benchmarks and are licensed for commercial use.
Llama 3 models include base and instruction-tuned models. You can chat with the instruct models right now at http://llama3.replicate.dev.
Llama 3 is made for developers. It’s permissively licensed and it comes in useful sizes. Use Llama 3 70B models for dialogue, instruction following, code generation, reasoning. Or Llama 3 8B models for summarization, translation and other cases where efficiency is everything.
LLama 3 70b Instruct
Try the full API playground, where you can test Llama 3 in detail and get all the documentation you need: meta-llama-3-70b-instruct.
Read related articles: