EmbeddingGemma 2
Freegoogle/embeddinggemma-2Embed · 768 dimensions · Oct 2026 · 1 upstreamsupstream上游
EmbeddingGemma 2 is an open multimodal embedding model built by Google DeepMind which maps text (incl. code), images, video, and audio inputs—and combinations thereof—into a single, unified 768-dimensional vector space. The model has 740M total parameters, combining a 270M parameter text model with modular vision (170M) and audio (300M) encoders. AnyRouter exposes the text path via /embeddings — the OpenAI-compatible body is a text `input`, so the multimodal encoders (text/image/video/audio natively, 8,192-token shared window) are not reachable from this endpoint. Runs on the user's own paired device via Ollama and the AnyRouter relay.
- max inputđầu vào tối đa最大输入
- 8K
- dimensionssố chiều维度
- 768
- inputđầu vào输入 / 1M
- $0
- routestuyến路由
- 1
Upstreams
1- 1OllamaRelayFree
Quick start
curl https://anyrouter.dev/api/v1/embeddings \
-H "Authorization: Bearer $ANYROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "google/embeddinggemma-2",
"input": "The quick brown fox jumps over the lazy dog"
}'Details
- Owner
- Kind
- Embeddings
- Released
- Oct 2026
- Context
- 8K
- Data policy
- Zero retention
- Endpoint
/api/v1/embeddings
- Search and RAG
- Deduplication
- Clustering
Upstreams and pricingUpstream và giá上游与价格


params: input · model · dimensions · encoding_format
Uptime + latency
No health checks recorded for this model yet.
Across all upstreams
Loading usage…