v0.40.0-rc6: model: add multimodal embeddings (#18820)
Implements the EmbeddingGemma2Model architecture on the MLX runner: 24-layer bidirectional text encoder with PLE, shared gemma4 vision/audio towers, mean-pool + L2…
What moved
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.
Why it matters
Filed as a public note from Ollama, 2026-10-06. The desk tags this brief Hardware.
On the record
- Filed from the Ollama official GitHub releases feed on 2026-10-06.
- Primary source host: github.com.
- Implements the EmbeddingGemma2Model architecture on the MLX runner: 24-layer bidirectional text encoder with PLE, shared gemma4 vision/audio towers, mean-pool + L2 output.
Primary source
github.com
github.com
Desk
Logged as brief 073
Logged as brief 073
Cite this brief
v0.40.0-rc6: model: add multimodal embeddings (#18820) (Ollama, 2026-10-06) — https://labledgerdesk.pages.dev/b/2026/10/06/v0-40-0-rc6-model-add-multimodal-embeddings-18820/