Release
FP4 Just Landed in llama.cpp: NVFP4 vs MXFP4 Explained (2026)
NVFP4 in llama.cpp, MXFP4 in ik_llama.cpp. The first practical FP4 quantization for the GGUF ecosystem — what works, what doesn't, and what to test.
DeepSeek V4 Flash vs Pro: What Actually Dropped and How to Run It
DeepSeek V4 preview dropped April 23 with two MoE variants: Pro at 1.6T/49B active and Flash at 284B/13B active. Both MIT, both 1M context. Flash is the news.
Best New Ollama 0.17 Features: ollama launch, MLX, and OpenClaw Support
Everything new in Ollama 0.16 through 0.17.7: ollama launch for coding tools, native MLX on Apple Silicon, OpenClaw integration, web search API, and image generation. Updated March 2026.