AI, automation & information security
Patrick Gawron
Agentic workflows, local-first AI, secure business logic. Daily AI experiments, consulting projects - and a build log of coding sidequests.
Tools
Notes
Quick posts for everything that doesn't need a full article - local LLMs, CSS, keto. One page each.
Browse all notes >Tech news
Anthropic releases Claude Fable 5.2 and Mythos 5.2, Fable tops Artificial Analysis at 65
API-only, no local weights. Artificial Analysis scores Fable 5.2 at 65, the highest index so far, which resets the bar that open-weight models are measured against. Worth tracking as the frontier reference for local releases.
MiniMax releases H3-Image, an open-weight image model built on H3
Open weights for an image model built on the H3 stack, on Hugging Face. Parameter count, licence and VRAM needs are not in the source text, so check the model card before trying it in ComfyUI on a 16 to 32 GB card.
DeepSeek-V4.1-Flash-1017 Q2 GGUF runs at 41 tok/s on one RTX 3090 on Windows
A Q2 GGUF of DeepSeek-V4.1-Flash-1017 hits 41 tok/s average on a 3090 with 192 GB of RAM, natively on Windows. The poster reports 90.2% next-token agreement with DeepSeek's API on general text and 92.4% on code/math. Self-reported.
Liquid AI releases the LFM3 family with GGUF builds and day-zero llama.cpp support
LFM3 spans a 1.7B text model with 256K context (about 1 GB at 4-bit, trained for agent harnesses), plus 2B omni, audio and VL variants. The Omni-2B GGUF shipped with a same-day llama.cpp release, so it runs on any card.