LiteLLM now supports gemini-3.8-flash on day 0, on both Google AI Studio (gemini/) and Vertex AI (vertex_ai/). Google calls its newest Flash model its primary general-purpose model, with gains over 3.7 Flash on long-horizon software engineering, agentic tasks, and multi-step reasoning, and on DeepSWE v1.1 it outperforms most larger frontier models at a fraction of the cost.
LiteLLM now supports gemini-3.7-flash on day 0, on both Google AI Studio (gemini/) and Vertex AI (vertex_ai/). Google's newest Flash model delivers faster responses with meaningfully better quality than 3.6 Flash and scores higher on complex multi-step agentic, coding, and reasoning benchmarks.
LiteLLM now supports gemini-3.5-flash with full day 0 support!
If you only want cost tracking, you need no change in your current LiteLLM version. But if you want support for new features introduced with this release, namely thinking levels, strict function-call IDs, and thought signatures, use v1.87.0-dev.1 or above.
LiteLLM now supports the Google AI Studio Managed Agents API. Create, manage, and run custom agents through LiteLLM.
Available from LiteLLM v1.87.0-dev.1 or above.
Litellm now fully supports Gemini Embedding 2 GA.
For end-to-end behavior, input shapes, and MIME types, see the Gemini Embedding 2 Preview walkthrough. This post focuses on GA naming, cost map coverage.
LiteLLM now supports gemini-3.1-flash-lite-preview with full day 0 support!
If you only want cost tracking, you need no change in your current Litellm version. But if you want the support for new features introduced along with it like thinking levels, you will need to use v1.80.8-stable.1 or above.
LiteLLM now supports gemini-3.1-pro-preview and all the new API changes along with it.
LiteLLM now supports gemini-3-flash-preview and all the new API changes along with it.
If you only want cost tracking, you need no change in your current Litellm version. But if you want the support for new features introduced along with it like thinking levels, you will need to use v1.80.8-stable.1 or above.
This guide covers common questions and best practices for using gemini-3-pro-preview with LiteLLM Proxy and SDK.
LiteLLM now supports multimodal embeddings with gemini-embedding-2-preview, mixing text, images, audio, video, and PDF content in a single request. Available via both the Gemini API (API key) and Vertex AI (GCP credentials).
gemini/...): each input element returns its own embedding, indexed 0..N-1, the same shape as OpenAI's /embeddings. LiteLLM routes to the batchEmbedContents endpoint with one EmbedContentRequest per input.vertex_ai/...): all input elements are combined into a single unified embedding via embedContent. Vertex AI does not expose batchEmbedContents for Gemini embedding models, so N parts → 1 vector. To get one vector per item, call embedding(...) once per input.