vLLM optimizations on a 6-GPU cluster reduced voice AI latency by 40% for a Qwen-based system, enabling 500 concurrent sessions per node without hardware upgrades.

Ollama integrates support for Codex with DeepSeek V4, Gemma 4, Qwen 3.6 for local execution

Benchmark revealed it collapsed under load of 5 concurrent users, highlighting gap between developer-friendly tools and production-ready systems.

Ollama expands its service to include cloud-hosted model deployment, starting with MiniMax's M2.7.

Added support for Apple's MLX framework as a backend for local LLM inference on macOS