Dev's model picks: Opus 5.5 as go-to, GLM for cheap fast prod, Astra for bug hunts
ivan_bezdomny · x · 2026-10-05
A developer shares practical picks across frontier coding models:
- Opus 5.5 is the go-to online model for most people, especially those who don't want multi-agent workflows or always-on prod jobs
- Astra is good at finding bugs and feature design but writes sloppy, incomprehensible code — use it for results, not maintenance
- Claude still makes nice UIs
- For fast, cheap prod jobs, an open-weights model like GLM or GLM-flash beats "older" frontier models on both reliability and cost
- Jev pairs well with both Astra and Opus 5.5 for designing decision flows with filtering, merging and evaluations
Related event: Developers Share Real-World Model Split: Opus 5.5 Plus GLM(3 posts)→
More from coding & agent
- DIY AI race engineer: local open-source models give live F1 telemetry coaching — coreweavedevs · 2026-11-20
- Vibe-coded 'Spores' demo: cursor-reactive particles via Antigravity + Gemini — AI_Andrew · 2026-11-20
- Dev Keeps Asking Codex 'Wait, How Did You Do That?' as Agent Surprises Him — yungcontent · 2026-11-20
- Dev deletes all ADRs: 49% superseded in days, codebase is now your source of truth — sloppenheimer · 2026-11-20
- Anthropic opens Claude Forward Deployed Engineer learning path, ~9.5 hours self-paced — leslysandra · 2026-11-20
- Google open-sources Sashiko, a Gemini agent that cut Linux post-merge fix rates 35% — rseroter · 2026-11-20