OpenClaw memory already works with self-hosted EmbeddingGemma, but lacks multimodal embeddings
steipete · x · 2026-10-07
osanseviero notes that OpenClaw's memory system can already use Google's EmbeddingGemma embeddings today: just point it at an Ollama or OpenAI-compatible endpoint and self-host via vLLM, SGLang, or similar. He's unsure whether OpenClaw supports multimodal embeddings, which would be needed for memory over voice or vision inputs.
steipete questions whether such a feature belongs in the core at all — it could be something anyone adds via a plugin, and doesn't match what users are loudly asking for.
Related event: OpenClaw Memory System Now Supports Self-Hosted EmbeddingGemma(3 posts)→
More from coding & agent
- Open-Source Project With 2,000+ Stars Adds 9 AI Animation Explainer Styles and Full Workflow — dotey · 2026-10-07
- ChatGPT Launches Meetings Plugin That Auto-Notes Calls and Feeds Context Into Codex — craigsdennis · 2026-10-07
- Luma events via MCP can't add custom form questions, forcing users back to the browser — aronkor · 2026-10-07
- "Your agents want to work in a monorepo" — and the Bazel joke that followed — vboykis · 2026-10-07
- MCP Events UX is painful for consumer apps, developer reports six-step activation flow — haltakov · 2026-10-07
- Tip: force GPT-Live to say exact phrases via a developer message in session.start's input — pbbakkum · 2026-10-07