---
type: "post"
id: "google-s-embeddinggemma-2-open-embedding-model-and-fab029"
slug: "google-s-embeddinggemma-2-open-embedding-model-and-fab029"
title: "Google s embeddinggemma 2 open embedding model and apache 2 0 licensing"
description: "Thesis: Apache 2.0 licensing for EmbeddingGemma 2 gives embedding-system operators an alternative to relying exclusively on a hosted proprietary model."
retrieval_nugget: "Thesis: Apache 2.0 licensing for EmbeddingGemma 2 gives embedding-system operators an alternative to relying exclusively on a hosted proprietary model."
published_at: "2026-10-06"
updated_at: "2026-10-06"
record_date: "2026-10-06"
date_kind: "published_at"
topics: ["google"]
source_urls: ["https://simonwillison.net/2026/Oct/6/hn-49983751"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "post:google-s-embeddinggemma-2-open-embedding-model-and-fab029"
status: "published"
visuals: []
editorial_provenance: {"schema_version":"newruntime-editorial-copy-v1","content_status":"source_grounded_final","final_copy_sha256":"sha256:92d91338325c4b8ea16c92be47de5934c85ab7c2cf10627e03c749e4785454bb","reviewed_at":"2026-10-06T21:00:57.209Z","source_evidence_count":1,"verified_claim_count":3,"site_analysis_schema_version":"newruntime-site-analysis-v1","site_object_kind":"field_note","observed_fact_count":3,"implication_count":2,"watch_condition_count":2,"related_record_count":0}
analysis: {"schema_version":"newruntime-site-analysis-v1","object_kind":"field_note","thesis":"Apache 2.0 licensing for EmbeddingGemma 2 gives embedding-system operators an alternative to relying exclusively on a hosted proprietary model.","observed_facts":[{"text":"Simon Willison says EmbeddingGemma 2 is under the Apache 2.0 license.","source_urls":["https://simonwillison.net/2026/Oct/6/hn-49983751"]},{"text":"He says embedding applications often calculate thousands or millions of vectors and store them for later comparison.","source_urls":["https://simonwillison.net/2026/Oct/6/hn-49983751"]},{"text":"He describes the possibility that a vendor may stop offering a model, leaving users to recalculate stored vectors.","source_urls":["https://simonwillison.net/2026/Oct/6/hn-49983751"]}],"mechanism":"When a model changes or disappears, vectors produced by different models may require recalculation to maintain a consistent index. Open weights can preserve an operator’s option to run that model independently or seek another provider.","why_now":"The release makes licensing and fallback options salient for teams choosing an embedding dependency.","implications":["Teams with large stored indexes should account for model continuity and re-embedding costs when choosing a provider.","Open weights may offer a fallback path, while hosted inference can still be the preferred operating model."],"evidence_boundary":"This record relies only on the linked commentary and does not independently establish those operational details.","watch_conditions":["Whether providers offer stable versioning or migration support will determine how much practical protection open weights provide.","Check the official release materials for model specifications, supported deployment paths, and terms of use."],"related_records":[],"new_branch_reason":"No prior New Runtime record is supplied; this begins a distinct record focused on EmbeddingGemma 2 licensing and embedding-index portability."}
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/posts/google-s-embeddinggemma-2-open-embedding-model-and-fab029/","markdown":"https://newruntime.com/posts/google-s-embeddinggemma-2-open-embedding-model-and-fab029.md","json":"https://newruntime.com/posts/google-s-embeddinggemma-2-open-embedding-model-and-fab029.json"}
---

# Google s embeddinggemma 2 open embedding model and apache 2 0 licensing

## Retrieval answer

Thesis: Apache 2.0 licensing for EmbeddingGemma 2 gives embedding-system operators an alternative to relying exclusively on a hosted proprietary model.

Thesis: Apache 2.0 licensing for EmbeddingGemma 2 gives embedding-system operators an alternative to relying exclusively on a hosted proprietary model. The supplied commentary describes a portability option, not proof that any particular migration is effortless.

Observed facts: Simon Willison says EmbeddingGemma 2 is under the Apache 2.0 license. He says embedding applications often calculate thousands or millions of vectors and store them for later comparison. He also describes the possibility that a vendor may stop offering a model, leaving users to recalculate stored vectors.

Mechanism: When a model changes or disappears, vectors produced by different models may require recalculation to maintain a consistent index. Open weights can preserve an operator’s option to run that model independently or seek another provider. This is the mechanism: Thesis: Apache 2.0 licensing for EmbeddingGemma 2 gives embedding-system operators an alternative to relying exclusively on a hosted proprietary model.

Why now: The source points to a current model release and highlights its license as relevant to deployment choices. Why now: The release makes licensing and fallback options salient for teams choosing an embedding dependency.

Implications: Teams with large stored indexes should account for model continuity and re-embedding costs when choosing a provider. Implications: Open weights may offer a fallback path, while hosted inference can still be the preferred operating model.

Evidence boundary: The source is a commentator’s post linking to the release; it does not provide benchmark results, detailed model specifications, hosting availability, or guarantees about migration costs. Evidence boundary: This record relies only on the linked commentary and does not independently establish those operational details.

Watch conditions: Check the official release materials for model specifications, supported deployment paths, and terms of use. Watch conditions: Whether providers offer stable versioning or migration support will determine how much practical protection open weights provide.

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- topic hub: [Google](https://newruntime.com/topics/google.md): Explore the google topic hub.
- collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)