EmbeddingGemma 2

Google DeepMind · released Oct 6, 2026

UnknownUSmultimodal

Openness taxonomy

access, licence, and disclosure — never a single open/closed binary

Weights availability: unknown
Licence: Apache 2.0
Licence category: permissive
Commercial use allowed: unknown
Derivatives allowed: unknown
Redistribution allowed: unknown
Field-of-use restrictions: unknown
Training code availability: unknown
Inference code availability: unknown
Architecture disclosure: unknown
Training data disclosure: unknown

openness confidence

The report says the model ships under Apache 2.0

Deployability

raw values as extracted — free text, never inferred from branding

Self-hosting: unknown
On-premise: unknown
Edge deployability: unknown
API availability: unknown
Regions available: unknown

Benchmarks

as reported — vendor-supplied unless otherwise noted

No benchmark data on this release yet.

National-security pathway & evidence

from the assessed source item, when one is linked

Impact: mediumConfidence: lowNat-sec: noneindependent reporting

Evidence limitations: The supplied source text is a brief single-source report and gives no independent evaluation, detailed modality list, or explicit statement about downloadable weights.

Google DeepMind Releases EmbeddingGemma 2, a 740M Open Multimodal Embedding Model Built on Gemma 4

Linked research papers

from paper_model_links

No linked research papers yet.