EmbeddingGemma already works with OpenClaw memory via Ollama or self-hosted endpoints
osanseviero · x · 2026-10-07
Osanseviero confirms EmbeddingGemma can already be used with OpenClaw's memory search by pointing to Ollama or an OpenAI-compatible self-hosted endpoint (vLLM, SGLang, etc.). OpenClaw's memorysearch supports many providers including a key-free local llama.cpp setup (0.3GB GGUF), with queryInputType/documentInputType for asymmetric embedding endpoints. Possible gap: multimodal embeddings (voice/vision memory) may not be supported yet.
Related event: OpenClaw Memory System Now Supports Self-Hosted EmbeddingGemma(3 posts)→
More from coding & agent
- Hot take: AGENTS.md and agent skills are just text files—put in whatever works — intellectronica · 2026-10-07
- Set Up Your GrokBot Like Hiring a New Employee: One Job, Minimal Access — alex_verem · 2026-10-07
- ChatGPT Can Now Subscribe to Netlify Events: Auto-Reply Forms, Summarize Deploys — thisiskp_ · 2026-10-07
- Superconductor adds Connectors: link Linear, Sentry, Figma once for whole-team coding agents — sergeykarayev · 2026-10-07
- Nous Research launches Hermes Index agent leaderboard, Claude Opus 5.5 tops at 63.31 — NousResearch · 2026-10-07
- Coinbase Ships Agent Trading Features: Templates, TWAP Orders, and MCP Access — kleffew94 · 2026-10-07