AI Briefing#apple-silicon
The latest AI and developer news about #apple-silicon, with the original source and a short summary.
The latest AI and developer news about #apple-silicon, with the original source and a short summary.
OpenRouter costs about 1/3 as much as Apple Silicon and is 2x faster, based on comparable models.
Ollama has switched from a GGML-based build to direct llama.cpp support.
It's an MLX-based local AI engine that supports inference up to 4.2x faster than Ollama on Apple Silicon.
Triggered by GitHub Copilot's billing shift, local LLM hardware is proposed as a solution for AI self-sufficiency.
A lightweight, Mac-exclusive coding agent built on Apple's MLX framework has been released.
An analysis of Hugging Face user data found that AI builders tend to prioritize VRAM capacity over compute performance.
Benchmark analysis results on LLM inference performance, power efficiency, and MoE model running performance across Apple Silicon chipsets have been released.
Analysis results of LLM inference performance and power efficiency for Apple M4/M5 and A18 Pro chips have been released.
MTPLX, an engine that leverages MTP heads to boost LLM inference speed by up to 2.24x on Apple Silicon, has been released.
Phosphene has released a local video/audio generation panel for Apple Silicon.
On an M2 MacBook Pro with 32GB RAM, Qwen 3.6 27B's processing speed plummeted to 4t/s at a 52,000-token context.
On an M5 Pro 64GB, Qwen 3.6 35B A3B was faster and more accurate than the 27B.
Hunyuan3D-Paint and Shape were ported to MLX/MPS for Apple Silicon, making texture generation 3x faster.
Ported Hunyuan3D's image-to-mesh/texture pipeline to Apple Silicon.
Ported TRELLIS.2 to run 3D generation on Apple Silicon without CUDA.
According to a Bloomberg report, the new Mac Studio could be delayed until October.
By sharing Wasm memory with the GPU, inference runs zero-copy on Apple Silicon.
On M3 Ultra, Qwen 3.6 35B achieved a 100% tool-calling success rate across most of the 5 agent frameworks.
Darkbloom, a decentralized AI inference network that cuts costs by 70% by leveraging idle Apple Silicon Macs, has launched.
PrismML has released Ternary Bonsai, a family of 8B, 4B, and 1.7B LLMs using 1.58-bit ternary weights.
A summary of the performance and limitations of local music generation on Apple Silicon by running ACE-Step 1.5 with MLX.
An image-to-SVG pipeline that runs the entire process locally, from Moondream to VTracer.
It links idle Apple Silicon Macs together to build an end-to-end encrypted inference network.
A fast, uncensored local model optimized from Gemma 4 26B into MLX 4-bit.