server-v0.0.5

OpenMOSS/MOSS-Transcribe-Diarizeserver-v0.0.5Jul 10, 2026by Dhravya

AI Summary

Adds support for pluggable embedding models, allowing self-hosted instances to use local ONNX, OpenAI, or Google models instead of a default local model.

Key Highlights

Pluggable embeddings for local ONNX, OpenAI, or Google models
Configuration via environment variables (`SUPERMEMORY_EMBEDDING_*`)
Safe upgrade mechanism for stores without locks

New Features

Pluggable embedding model support
OpenAI and Google embedding integration
Safe upgrade handling for legacy stores

Full Release Notes

Self-hosted supermemory can now use local or remote embedding models.

- **Pluggable embeddings** — local ONNX (default) or OpenAI / OpenAI-compatible / Google via `SUPERMEMORY_EMBEDDING_*`; first-boot picker + plan lock in `embedding-plan.json`
- **Safe upgrades** — populated stores without a lock assume legacy `local · Xenova/bge-base-en-v1.5 · 768d`; same-dimension model switches fail fast instead of mixing vectors

```sh
curl -fsSL https://supermemory.ai/install | bash
```

```sh
supermemory-server upgrade
```