What's Changed * openai: allow Responses effort on non-thinking models by @hoyyeva in https://github.com/ollama/ollama/pull/19010 * llama-server: let llama-server infer the draft speculative type by @dhiltgen in https://github.com/ollama/ollama/pull/19064 * mlx: add PPLX Decider v1.2 support by @dhiltgen in https://github.com/ollama/ollama/pull/19072 **Full Changelog**: https://github.com/ollama/ollama/compare/v0.40.6...v0.40.7-rc0