Apple's upcoming M5 Ultra, with 512GB memory and 1.2TB/s bandwidth, is projected to cost $15,000–17,000 when it ships in October, making local AI inference for models like DeepSeek V4 Pro, Kimi K3, an…
Apple's 2026 Mac mini, introduced this week with the new M6 and M5 Pro chips, offers up to four times faster AI performance and twice-as-fast graphics versus the 2024 M4 model, though ports, storage, …
An engineer found that their local RAG app was slow because it re-read every retrieved document from token zero on every question, with prefill dominating generation by roughly five to one. The fix in…
Apple launched a new Mac mini with M6 and M5 Pro chips, featuring up to 4x faster AI performance, 2x faster storage and graphics, and 40 percent faster CPU performance compared to M4 models. The M6 mo…
Darkbloom, a service at darkbloom.dev, lets owners of M-Series Macs rent out idle compute to serve open-source language models, with estimated earnings of $146 to $602 per month depending on chip and …
Apple's M4 Pro with 64GB RAM can run vision models like Llama 3.2 Vision, Moondream2, and Qwen2-VL locally with low latency, according to a developer guide. The guide recommends using Ollama to manage…
A developer explains that pure CPU-only local inference performance is fundamentally limited by memory bandwidth, not CPU cores, and provides a formula to calculate the theoretical maximum tokens per …
Singapore startup Acrab announced its GELIX 1 edge AI system-on-chip and Agent Box reference system on July 23, targeting local inference for models in the 100-billion-parameter class. Acrab says the …
A developer reports on two months of real-world testing of a local AI development stack, finding that performance is limited to 4-5 tokens per second and that memory constraints cause frequent crashes…
A bug report on OpenAI's Codex GitHub repository claims that the desktop app's "thinking" animation causes excessive GPU usage, with a user reporting up to 70% utilization and 1.5 W power draw on an M…
Two of China's largest AI model platforms, Doubao and Qwen, tightened restrictions on custom agents last week, delisting or throttling third-party agents and narrowing creation permissions. The develo…
Apple raised the starting price of the M4 Pro Mac Mini by $200 to $1,599, citing rising memory and storage chip costs driven by AI data center demand. The company also increased prices across Mac and …
Beelink's SER10 Max mini PC, powered by AMD's Ryzen AI 9 HX 470 "Gorgon Point" chip, matches Apple's M4 Pro on a single TypeScript benchmark but falls flat elsewhere, with local LLM performance hamper…
Apple's Mac mini with M4 chips is facing severe shortages due to high demand from AI developers and component constraints, with wait times of up to three months. The company is expected to refresh the…
NVIDIA's RTX 5090, priced at $1,999, offers 32 GB of GDDR7 memory with 1,792 GB/s bandwidth, making it the fastest consumer GPU for both gaming and local AI inference. The card excels at running 32B-c…
Apple's Mac Studio with M3 Ultra, offering up to 512 GB of unified memory at 819 GB/s, is the most practical desktop solution for running large local AI models like 70B to 400B-class quants without mu…