Ollama 0.32 turns the CLI into an agent — with a cloud model as the default

Ollama v0.32.0 makes plain `ollama` launch an interactive coding/delegation agent, sets GLM-5.2:cloud as the default chat/code model, and adds deprecation warnings for older local agent models (CodeLlama, Qwen2.5, Llama 3.x, Mistral, StarCoder, base DeepSeek-R1). The flagship local-model runner is repositioning as an agent front-end that mixes local and cloud.

What it means If your self-hosted stack shells out to `ollama`, the default behavior just changed — and the deprecation list says which local models to migrate off.