A GTM engineer's 30 days of Cursor: model 5.6 Sol takes the biggest token share
brandon_galang · x · 2026-08-27
A GTM engineer shares 30 days of real @cursorai usage: most dev work runs through @can1357's oh-my-pi CLI harness, dropping back into the GUI only for visual inspection or more general agent tasks.
Token breakdown: 5.6 Sol (medium) > Grok 4.6 (medium fast) > Gemini 3.7 Flash (medium) > Claude Fable 5 (low) > GPT 5.6 Luna (medium) > others. Gemini 3.7 Flash used to absorb most chat/planning tokens, but as work demanded more precision, he cut over to 5.6 Sol.
More from coding & agent
- Vercel open-sources vgpu, a WebGPU library designed for agents — evilrabbit_ · 2026-08-27
- New Tool Runs Codex, ClaudeCode, OpenCode, or Pi Side-by-Side — tom_doerr · 2026-08-27
- DocETL: New Python Library for Agentic Data Processing and ETL with AI — mdancho84 · 2026-08-27
- Beyond the model: what's in your agent setup? — External-Wind-5273 · 2026-08-27
- Codex users would trade 5% quota to let running tasks finish — vinmi · 2026-08-27
- qmax-mcp: local QA MCP server that scans URLs and runs Playwright repros for coding agents — bestofdesp · 2026-08-27