From the Shorts
Companion guides for my @avnish.codes shorts — the links, commands and cheat sheets.
5 Chinese AI Models You Can Download Right Now (Open Weights)
Five open-weight Chinese LLMs you can download today — Qwen3.8-Flash-Next, MiniMax-M3, GLM-5.3, DeepSeek-V4-Pro and Kimi K3 — with the official Hugging Face links.
02 · BlogRemoveMacAI: remove Apple Intelligence on macOS 27
A Swift CLI by omlahore that disables Apple Intelligence daemons, clears cached on-device models and returns the disk space — with an undo line printed before it exits.
03 · BlogLLMs vs AI agents vs workflows: which one are you building?
An LLM labels the intent. A workflow routes it down a fixed path. An agent picks a tool and loops. Here's how to tell which one you're actually building.
04 · BlogApex Governor Limits: Why Query 101 Kills Your Transaction
Sync Apex gets 100 SOQL queries, 150 DML statements, 10 seconds of CPU and 6 MB of heap per transaction. Cross the line and the whole transaction rolls back.
05 · BlogWhat Is a Token in an LLM? Tokens vs Words, Explained
The model never reads your words — a tokenizer cuts text into chunks first. Here's how tokens drive your API bill, context window and rate limits.
06 · BlogCalling an LLM From Apex: Where Python Still Wins
Apex makes the HTTP call to an LLM in about twenty lines. Everything that decides whether the feature survives — validation, retrieval, evals, cost — is the part Python still owns.