A side-by-side comparison of Gemini Embedding 001 and Gemini Embedding 2 Preview: input/output pricing, capabilities and available endpoints, served live from openbat.ai. Both are reachable with the same API key.
| Item | Gemini Embedding 001 | Gemini Embedding 2 Preview |
|---|---|---|
| Input (per 1M tokens) | $0.1125 | $0.15 |
| Item | Gemini Embedding 001 | Gemini Embedding 2 Preview |
|---|---|---|
| Context window | 2,048 | 8,192 |
| Max output | 2,048 | 8,192 |
| Item | Gemini Embedding 001 | Gemini Embedding 2 Preview |
|---|---|---|
| multimodal | — | Yes |
Gemini Embedding 2 Preview does — Gemini Embedding 001 accepts 2,048 input tokens and Gemini Embedding 2 Preview accepts 8,192.
only Gemini Embedding 2 Preview supports multimodal.
Yes. Both are available on openbat.ai through one API key and the same OpenAI-compatible endpoint — switching means changing the model field from "gemini-embedding-001" to "gemini-embedding-2-preview", nothing else.
Both are available on openbat.ai under one API key — switching between them means changing the model field and nothing else, so you can use each where it fits rather than picking one.
Gemini Embedding 001 · Gemini Embedding 2 Preview · All models