AI Coding + Performance
Posts matching all selected tags.
Category
All Categories
Tags
6 more tagsFewer tags
Newsletter
New posts land in your inbox when they publish. No spam, unsubscribe anytime.
You're in. New posts head your way.
10 posts
Grok 4.6 tied Sol and still lost the terminal
Grok 4.6 Terminal-Bench is the row that should move the buy. It tied Sol at 61 on the index and still sat at 26 percent on v3.0. Shop a named agent loop.
750 tokens a second does not make review faster
GPT-5.6 Ultrafast is 750 tok/s of author speed. That fills the human review queue harder. Pin the changelog to the bottleneck, then budget the review lane.
OpenCode vs Claude Code on the Same Tasks: Same Passes, Different Paths
OpenCode vs Claude Code on four graded tasks. Both hit 4/4. Paths diverged on tools, wall time, and reported tokens. Product lane, not a same-model lab.
Fat Instruction Files Hurt Frontier Coding Agents
A fat instruction file costs more and can follow less. Thin rules passed every check. Fat files dropped to 80% and burned 21x the prompt tokens on the same task.
Independent verification of Bun memory usage claims
Bun's Rust rewrite claims lower memory usage. A firsthand build-loop test on the exact version Bun benchmarked checks whether those numbers actually hold up.
Claude Code vs Codex Limits, Measured in the Same Unit
Claude Code vs Codex limits, measured across 28 days and 56.8 billion tokens. Why token counts cannot predict quota burn, and what to read on each side instead.
Why expensive models make terrible execution subagents
Subagent model selection has one costly default. Reasoning models obey their own traces 15.6% of the time, so here is where to pin model and effort.
Token-Compression Skills Do Not Survive Measurement
Token compression skills report savings the invoice never shows. Cache reads carry 87 percent of an agent bill, and cutting tokens can push the cost up.
7 Free Ways to Stop ChatGPT From Eating Your CPU, RAM, and Disk
Fix ChatGPT performance without spending a cent: diagnose high CPU, reduce RAM-heavy threads, stop disk churn, trim Git load, and disable unused MCP servers.