# Ollama v0.40.4 — Release & Upgrade Intelligence
- **Product:** Ollama (https://ollama.com)
- **Version:** v0.40.4
- **Released:** Oct 12, 2026
- **Impact Score:** 5/10
- **Breaking Changes:** Not verified
- **Canonical Analysis:** https://changelogs.us/ollama/updates/v0.40.4/

## Executive Summary (TL;DR)
Ollama v0.40.4. Source highlight: Decision models in the CLI

## New Features
- Added support for multimodal embeddings through llama-server

## Improvements
- Improved inference performance by reusing verified manifests and already loaded models
- Improved Gemma 4 prefill performance on NVIDIA GPUs with the MLX runner
- Improved speculative decoding reliability in the MLX runner, including recurrent state handling after rollback

## Bug Fixes
- Fixed loading multimodal embedding models with small context windows