Ollama v0.40.16 adds Ripple Preview to its MLX runner
Ollama v0.40.16 added Ripple Preview support to its MLX runner, allowing users to launch Ricursive Intelligence's 65B dense model for agentic coding with `ollama run ripple-preview`. The release also reports that Qwen3.8 Flash Next uses about 27 GiB less memory on Apple silicon, though no independent measurement was supplied. Other changes fix Claude Code routing, clarify retired cloud-model errors, forward truncation options for the OpenAI Responses API, and return an empty array from `/v1/models` when no models are installed.
The update makes a 65B agentic-coding model available through Ollama's MLX runner while improving local inference and Claude Code workflows.
Ripple Preview
Ripple Preview from Ricursive Intelligence is a 65B dense model built for agentic coding.
ollama run ripple-previewWhat's Changed
- Qwen3.8 Flash Next uses about 27 GiB less memory on Apple silicon
ollama launch claudealways routes Claude Code to the selected modelollama launchexplains when a cloud model has been retired/v1/responsesrespectstruncationand no longer silently truncates by default/v1/modelsreturns an empty list instead ofnullwhen no models are installed- CUDA 13.5 support for Windows on ARM
- Updated llama.cpp
Full Changelog: v0.40.15...v0.40.16
#Ollama#Ripple Preview#Ricursive Intelligence#MLX#Qwen3.8 Flash Next#Claude Code#OpenAI Responses API