Sharing a Multi-Model Coding Workflow
gregmushen · x · 2026-07-12
The author shares their currently effective coding workflow:
- Using Opus 4.8 for planning
- Using Grok Build / Grok 4.5 to write code according to the plan
- Creating PRs based on the results
- Using Sol 5.6 extra-high for PR review (which they feel might be a bit overkill)
- Using Grok Build / Grok 4.5 again to implement review feedback
- Repeating the review and revision loop
They added that they are also using 5.6 luna extra-high for development, which performs as well as or better than 5.5; however, Grok Build / Grok 4.5 is faster and performs similarly to 5.5 in the review loop.
Related event: Multi-Model Coding Workflow Splits Planning and Implementation(2 posts)→
More from coding & agent
- Kolab Benchmark Uses Pro Screen Recordings to Show AI Agents Overestimate Collaboration Quality — minnesotanlp · 2026-10-27
- Dev argues Claude burns far more tokens than GPT models for the same coding work — haider1 · 2026-10-27
- Stop building agents that can't change models: stable contracts, external state, real evals — rseroter · 2026-10-27
- Cheap models can cost more in agent harnesses than frontier ones, researcher warns — vasuman · 2026-10-27
- Kanzo: A 5k-Line Rust Kanban Binary With MCP for Claude Code and Codex — allmudi · 2026-10-27
- 'A human approves everything' is a trap: approval fatigue and how to actually design safe agents — Efistoffeles · 2026-10-27