First Google’s TurboQuant and then this repo: https://github.com/danveloper/flash-moe Innovations like these will make local hosting more and more practical.

https://www.youtube.com/watch?v=Kaq5Ual2ij8 @Stephen King gonna try it later, setting up something else for my local more secure info set up.

I'm looking at this right now. It super new, but how secure, will it go virial, will it get as much support? https://www.youtube.com/watch?v=EHlqRx0r4BI

Things got a kickstarter, but I'm good for now. Kinda wish I didn't buy the ram I needed, have to make income from this before I start buying more toys :P https://www.youtube.com/watch?v=RkzCAaIV_cQ

Go this email and tried it out.. really workable and the speed is good. ollama launch claude --model qwen3.5:35b-a3b-coding-nvfp4 email from Ollama.com Today, we're previewing the fastest way to run Ollama on Apple silicon, powered by MLX, Apple's machine learning framework. This change accelerates your most demanding work on macOS: Personal assistants like OpenClaw Coding agents like Claude Code, OpenCode, or Codex Fastest performance on Apple silicon, powered by MLX Ollama on Apple silicon is now built on top of Apple’s machine learning framework, MLX, to take advantage of its unified memory architecture. This results in a large speedup of Ollama on all Apple Silicon devices. On Apple’s M5, M5 Pro and M5 Max chips, Ollama leverages the new GPU Neural Accelerators to accelerate both time to first token (TTFT) and generation speed (tokens per second). Testing was conducted using Alibaba’s Qwen3.5-35B-A3B model quantized to NVFP4 and Ollama’s previous implementation quantized to Q4_K_M using Ollama 0.18. Ollama 0.19 will see even higher performance (1851 token/s prefill and 134 token/s decode) when running withint4 quantization NVFP4 support: higher quality responses and production parity Ollama now leverages NVIDIA’s NVFP4 format to maintain model accuracy while reducing memory bandwidth and storage requirements for inference workloads. As more inference providers scale inference using NVFP4 format, this allows Ollama users to share the same results as they would in a production environment. It further opens up Ollama to have the ability to run models optimized by NVIDIA’s model optimizer. Other precisions will be made available based on the design and usage intent from Ollama’s research and hardware partners. Improved caching for more responsiveness Ollama’s cache has been upgraded to make coding and agentic tasks more efficient. Lower memory utilization: Ollama will now reuse its cache across conversations, meaning less memory utilization and more cache hits when branching when using a shared system prompt with tools like Claude Code.