Try out Llama3.1-405B for free on Wirepod

You can explore the latest open source model from Meta without providing a credit card

In our last post, we discussed how you can use many OpenAI API compatible Large Language Model (LLM) service providers to serve Vector robot’s knowledge graph and intent graph. This post is a continuation of the previous post… we will explore how to use the best performing openly available LLM: Meta’s Llama 3.1 405B for FREE.

Llama 3.1 405B

Llama3.1 405B is currently the best openly available LLM. At the time of its release in July 2024, Meta stated that Llama3.1 405B was comparable with leading proprietary models such as GPT-4, GPT-4o, and Claude 3.5 Sonnet in close to 150 benchmarks that they ran tests on. Many leading LLM providers such as the ones we discussed in our previous post: Together.ai, Fireworks.ai, and Lepton.ai offer APIs for Llama3.1 405B. The typical rate for Llama3.1 405B at the time of writing this article is USD $2.80 per million input and output tokens, although that varies between service providers. While that is a cheap rate, I can see many students, hobbyists and…