Implements the EmbeddingGemma2Model architecture on the MLX runner: 24-layer
bidirectional text encoder with PLE, shared gemma4 vision/audio towers,
mean-pool + L2 output. /api/embed accepts per-item media via input dicts.
Intelligence View
⚡ tsecurity.de Intelligence
GitHub Release: ollama/ollama v0.40.0-rc6 (06.10.2026)
Release v0.40.0-rc6 von ollama/ollama auf GitHub: Implements the EmbeddingGemma2Model architecture on the MLX runner: 24-layer bidirectional text encoder with…
Intelligence Digest — kostenlos Täglich die wichtigsten Security-News · jederzeit abbestellbar