EmbeddingGemma 2: an open, lightweight multimodal embedding model
deepmind.google·6d ago
AI Engineering
Hacker News·8d ago·5 min read
Run Qwen 3.8 Flash Next (125B) on consumer hardware (RTX 4090) at 100T/s
TL;DR
Qwen 3.8 Flash Next, a 125-billion-parameter AI model, can now run on consumer hardware like the RTX 4090, enabling advanced tasks without server reliance. This breakthrough allows users to chat, code, and analyze images directly on their PCs.
✦ Why It Matters
Download and run Qwen 3.8 Flash Next on your RTX 4090 to explore advanced AI capabilities today.
Key Takeaways
- Install Strata to run Qwen 3.8 on your PC.
- Ensure your hardware meets the minimum requirements for optimal performance.
- Utilize the Strata app for real-time AI interactions.
- Experiment with different model sizes based on your RAM capacity.
How It Works
Strata operates by distributing the workload of the Qwen 3.8 model across the PC's resources. The graphics card handles the most frequently used components of the model, while RAM stores additional data, and the SSD serves as a backup for larger datasets.
This architecture allows the model to function efficiently on consumer hardware, making advanced AI accessible to a broader audience.