I use Pi coding agent with OpenAI GPT-5.6 Sol but have been finding that GPT-5.6 is slower than Claude Code. Do you use any other model like one of the hosted OSS ones?

I think GPT 5.6 Luna works great honestly for most boring coding tests. And saves money.

GPT-5 / GPT-5.5 Claude Opus / Sonnet Gemini Kimi K2 DeepSeek Grok

Do you pay for the API with Opus? since i think we can't use subscriptions with non claude code agents

GLM 5.x coding plan from z.ai, and opencode go plan.