SevenTnewS

MLX

5 published articles

Meta AI3 min read

Open Source AI

Meta's 30B Muse Glimmer hits Apple Silicon first, NVIDIA support follows

Meta released Muse Glimmer, a 30B open-weights multimodal model built for agent workloads, and Ollama shipped it for Apple Silicon the same day. DFlash makes it 1.5x to 1.8x faster on Mac hardware. NVIDIA and AMD support follow in the coming days.

2026-08-13

AndroidFeatured1 min read

Mobile AI

Gemma 4 goes fully offline on mobile, no cloud required

React Native developers can now embed Gemma 4 for offline inference with hardware acceleration on both Android and iOS. The model handles vision and tool-use tasks locally, as demonstrated by reading a flyer and scheduling a calendar event entirely on-device.

2026-07-09

Benchmarks & TestsFeatured3 min read

Performance

Gemma 4 runs 90% faster in Ollama 0.31 with a trick that needs no config

Ollama 0.31 introduces multi-token prediction for Gemma 4 on Apple Silicon, achieving near 90% faster token generation on coding benchmarks. The speedup comes from an auto-tuned draft model and a custom MLX kernel that eliminates redundant weight reads.

2026-06-29

VibeCodingFeatured3 min read

Apple Silicon update

Ollama's Mac update just solved the three things that made local AI painful

Ollama's MLX engine update brings NVFP4 quantization for higher quality 4-bit models, 20% faster inference through fused Metal kernels, and a snapshot caching system that eliminates reprocessing of shared context in multi-agent, thinking-model, and branching scenarios.

2026-06-11

AIFeatured4 min read

Local AI Inference

Ollama rewrote its Apple engine. The M5 results are finally competitive.

Ollama previews MLX-backed inference on Apple silicon, pushing Qwen3.5-35B-A3B to 1,851 tok/s prefill on M5. The shift brings NVFP4 quantization for production parity and a smarter cache for agentic workflows.

2026-03-30