EmbeddingGemma 2
- ID
- 32479
- Status
- summarized
- Published
- 07 Oct 2026, 4:37 AM
- Fetched
- 07 Oct 2026, 4:57 AM
- Provider
- Simon Willison
- Category
- developer-ai
- Original URL
- https://simonwillison.net/2026/Oct/6/hn-49983751/
- Source URL
- https://simonwillison.net/atom/everything/
Summary
- Score
- 6.5
- Created
- 07 Oct 2026, 4:57 AM
- Tags
- Audience
- developersai_ml_learnersdatabase_learnerssaas_startup_founders
What happened
Simon Willison comments on EmbeddingGemma 2 being under Apache 2.0, arguing that embedding models should not be closed, hosted-only services because apps store thousands to millions of vectors and a vendor deprecation can force costly re-embedding. He notes OpenAI once offered to cover re-embedding costs in April 2024 but says that cannot be relied on, and says he prefers paying a hosted provider while knowing he can fall back to open weights or another vendor.
Why it matters
If you build RAG or semantic search, this is a warning to pick embedding models with an open-weights fallback or multiple hosts, because a model retirement can turn into a full re-embedding bill across your stored vector corpus. EmbeddingGemma 2's Apache 2.0 license gives one such fallback path, but the text gives no benchmarks, pricing, or migration tooling, so it is not a performance or cost recommendation.
Discussion angle
How would you design a vector store so embeddings can be swapped without re-embedding everything—model-versioned collections, dual-write, or paying a provider that still gives you an open-weights escape hatch?