What’s New in Ollama 0.40.16: Ripple Preview 65B, and a Big Apple Silicon Memory Cut

Ollama 0.40.16 is out. Here’s what stood out, after 0.40.15.

• Ripple Preview from Ricursive Intelligence is a 65B dense model built for agentic coding: ollama run ripple-preview.
• Qwen3.8 Flash Next uses about 27 GiB less memory on Apple silicon. That’s a huge deal if you’re close to the unified-memory ceiling.
• ollama launch claude always routes Claude Code to the selected model, and ollama launch explains when a cloud model has been retired.
• API fixes: /v1/responses respects truncation instead of silently cutting off, and /v1/models returns an empty list instead of null when nothing is installed.
• CUDA 13.5 support for Windows on ARM, and an updated llama.cpp.

My take

The memory drop is the practical win. I’ve written up a plan to try Ripple Preview and measure that Qwen change.

Review before you install: read the release notes yourself, back up your config, and try new versions somewhere you can afford to break first.

God Bless, and May the Source Be with You!

References & Further Reading

• Ollama v0.40.16 (GitHub)
• What’s New in Ollama 0.40.15 (EtherNinja)