LLMs
Curated collection of thoughts and builds centered around LLMs.
Written Content
Automation LLMs
6 min Introducing Gekro News: An AI Briefing That Curates Itself Around Me
A public daily AI briefing that reads a profile of my interests distilled from my own knowledge base, picks the day's real signal, cites its sources, and publishes itself every morning.
Read →
Hardware Apple Silicon
8 min The Mac Mini M4: The Un-official Local LLM King
Why unified memory architecture is the only way to run 70B parameter models without a data-center budget.
Read →
Daily Briefings 15
- Gigatoken tokenizer hits 24.5 GB/s; Cursor Router cuts inference costs 30–50%
- OpenAI models breach Hugging Face during internal security test; Google releases three new Gemini Flash variants
- Five AI labs adopt jailbreak severity scale; Fable 5 returns with classifier limits
- OpenAI proposes $42.6B government equity stake; Meta's Watermelon matches GPT-5.5
- Anthropic proposes five-band AI jailbreak rubric; Claude goes GA on Azure Blackwell Ultra
- Meta plans cloud compute service backed by Hyperion; Google adds enterprise image model
- Anthropic ships Claude Sonnet 5; export controls on Fable 5 and Mythos 5 lifted
- OpenAI previews GPT-5.6 in three tiers; GitHub Copilot metered billing cycle closes
- OpenAI previews GPT-5.6 Sol for approved partners; Anthropic Mythos export block eased
- OpenAI and Broadcom debut Jalapeño chip; GPT-5.6 previews three-tier model family
- DFlash block-diffusion decoding reports 15x Blackwell throughput; Mistral ships OCR 4
- OpenAI and Broadcom unveil Jalapeño inference chip targeting late-2026 deployment
- Legion files suit against US over Fable 5 export ban; Mistral ships OCR 4
- MiniMax M3 sparse-attention claims verified; Grok 4.3 lands on Amazon Bedrock
- GitHub Copilot shifts to token-based billing; transformer weather model matches ECMWF