Ollama v0.30.8 Release Focuses on Stability and KV Cache Reuse

Core summary

One sentence to understand this update

Ollama v0.30.8 addresses launch provider issues, enhances prompt caching by separating it from context shift for better KV cache reuse, and improves MLX inference stability.

Impact & opportunity

What this could mean

Developers using Ollama will experience more reliable model launching, improved performance due to efficient KV cache management, and enhanced stability for MLX inference.