2026-08-12 BREAKTHROUGHS☾ PM
Meta Unleashes Llama 3.1 405B. It Matches GPT-4. For Free, Obviously.
📰 THE BRIEF
Meta has released Llama 3.1 405B under a commercial license, making it the largest openly available model to date. Small teams can now download it and run it locally or through inexpensive cloud providers, achieving GPT-4 level performance without API fees or usage limits.
💡 WHY IT MATTERS
This demonstrates the principle of capability diffusion. The mechanism is open-weight licensing: when a frontier model is released with commercial rights, it compresses the gap between well-funded labs and resource-constrained developers. The lesson is that access, not just capability, determines who captures value from AI.
👥 WHO'S DOING IT
Meta released the model. The source links to the official announcement at ai.meta.com/blog/meta-llama-3-1.
⚡ TRY IT
- Visit huggingface.co and search for Llama 3.1 to see the available model sizes and license terms. You will find downloadable weights and documentation.
- Sign up for a free account at a cloud GPU provider like Together AI or Groq, which offer hosted Llama models.
- Select Llama 3.1 from the model menu and send a test prompt through the provider's playground interface. You should receive a response comparable to GPT-4 quality without paying API fees.