My LLM Tier List
2026-02-05
A sourced summary of Daniil Sedov's note about keeping a personal, time-indexed ranking of frontier AI models based on real usage.
Posts
Short AI notes, updates and ideas.
Things got too easy with AI
2026-01-27
A sourced summary about how AI can make small goals too easy, so users may need to aim higher.
I gave Codex its own Mac Mini
2026-01-09
A sourced summary about giving an AI coding agent its own computer and controlling it through a simple loop and Telegram bot.
TON Vanity: 286,000x faster vanity addresses
2026-01-02
A sourced summary of a technical post about building a much faster TON vanity-address generator with AI-assisted optimization.
AI in 2026
2025-12-31
A sourced summary of predictions about AI progress, agents, coding, science, companies, adoption, and risk in 2026.
What LLM to use today?
2025-11-28
A sourced summary about choosing AI models based on real tasks instead of only benchmark numbers.
There is nothing out-of-distribution
2025-11-20
A sourced summary of an argument that LLM generalization failures are often intelligence limits rather than pure data-boundary limits.
There is no singularity
2025-10-13
A sourced summary of the argument that AI progress may feel dramatic but still follow a smooth curve of technological acceleration.
Writing with AI
2025-10-12
A sourced summary about using AI for feedback and grammar help without letting it replace your own thinking and writing voice.
My impression of GPT-5
2025-08-11
A sourced summary of one user's experience with GPT-5, including expectations, early disappointment, coding use, routing issues, and instruction-following strength.
The complexity threshold of AI
2025-07-29
A sourced summary about why AI coding agents work better when large projects are broken into clear, low-complexity steps.
Billions of Tokens Later: Scaling LLM Fuzzing in Practice
2025-07-18
A sourced summary of large-scale experiments using LLMs for black-box and white-box fuzzing of compilers and related projects.
Multitasking in 2025
2025-03-31
A sourced summary about AI making multitasking easier by letting people delegate parts of their cognitive workload.
Documentation-Driven Compiler Fuzzing with Large Language Models
2025-03-26
A sourced summary of an experiment where LLM agents used documentation and a compiler binary to find real compiler issues.
Measuring and Analyzing Entropy in Large Language Models
2025-03-11
A sourced summary of a benchmark study on how random or predictable different LLM outputs are under number-generation prompts.