The world of local AI moves fast, and we’re moving right along with it. We’ve just pushed a major update to the LLM VibeCheck model almanac to include the latest and greatest models that have recently hit the scene.

Our goal is to ensure VibeCheck always gives you the most relevant recommendations for your hardware. Here’s a quick rundown of what’s new in this update:

• 🚀 The Llama 3.1 Family: Meta’s powerful new update is here. We’ve added both the 8B and 70B versions of Llama 3.1, which offer improved performance and efficiency over their predecessors.
• 💎 The Gemma 2 Invasion: Google’s next-generation open models have arrived. The new almanac now includes the highly anticipated Gemma 2 9B and its powerhouse big brother, the 27B model.
• 🐉 The Power of Qwen 2: We’ve added the latest high-performers from Alibaba’s Qwen team, including the very capable Qwen 2 7B and the massive 72B models.
• 👁️ Phi-3 Vision is here: Multimodality is the future. We’ve added Microsoft’s Phi-3 Vision, a powerful model that can understand both text and images, opening up a new world of possibilities for your local projects.

The updated list is already live. Simply launch LLM VibeCheck, and it will automatically fetch the new almanac, giving you fresh recommendations based on your hardware.

Stay tuned for more updates, and happy vibe-checking!

Follow the Vibe