{"slug": "gemini-2-5-models-extended-indefinitely-for-aactive-users", "title": "Gemini 2.5 models extended indefinitely for aactive users", "summary": "Google limited Gemini 2.5 model access in its API on September 18, 2026 to users who have actively used the models in the past, stating the 2.5 models are not deprecated and will continue to be served until further notice, while directing new projects to Gemini 3.5 Flash-Lite or Gemini 3.8 Flash. The same update log records the September 17, 2026 release of the antigravity-preview-09-2026 agent, which deprecates antigravity-preview-05-2026 and changes built-in tool parameters to PascalCase and file edits to line-range replacements, with the older agent shutting down on October 5, 2026. Google also made Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking generally available on September 15, 2026, Lyria 3.5 generally available on September 3, 2026, and Gemini 3.8 Flash generally available on September 2, 2026.", "body_md": "This page documents updates to the Gemini API.\n\n## September 18, 2026\n\n- **Gemini 2.5 models access update** : To ensure reliable performance for\neveryone, we are limiting access to the 2.5 models to users who have\nactively used them in the past. These models are not deprecated and will\ncontinue to be served until further notice through the API. For any new\nprojects, use our latest models: 3.5 Flash-Lite or 3.8 Flash. This\nhelps us maintain sufficient capacity for both ongoing legacy workflows and\nnew applications.\n\n## September 17, 2026\n\n- **Antigravity Agent 09-2026** : Released`antigravity-preview-09-2026` ,\nwhich replaces and deprecates`antigravity-preview-05-2026` .If you run on a remote sandbox ( `environment: \"remote\"` ) and read only`output_text` or`model_output` steps, update the agent string and nothing\nelse changes.If you run tools locally ( `local_environment` ) or parse`function_call` steps, the built-in tools changed. Parameters use PascalCase instead of\nsnake_case, and file edits use line-range replacements instead of full\nrewrites.Capability 05-2026 09-2026 File creation `write_file(path, content)`` write_to_file(TargetFile, CodeContent, Overwrite, Description)`File editing `write_file(path, content)` , full rewrite`replace_file_content(TargetFile, StartLine, EndLine, TargetContent, ReplacementContent)`File reading `read_file(path, offset, limit)` , byte offsets`view_file(AbsolutePath, StartLine, EndLine, ContentOffset)` Directory listing `list_files(path)`` list_dir(DirectoryPath)`File and code search None, agents used shell commands `find_by_name(SearchDirectory, Pattern, MaxDepth)` and`grep_search(SearchPath, Query, IsRegex)` Shell execution `code_execution(command, timeout_seconds)` Unchanged Web search `google_search(queries)` Unchanged See the [Antigravity Agent](https://ai.google.dev/gemini-api/docs/antigravity-agent) guide.`antigravity-preview-05-2026` shuts down on October 5, 2026, tracked on the[deprecations](https://ai.google.dev/gemini-api/docs/deprecations#managed-agents) page.\n\n## September 15, 2026\n\n- **Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking generally available\n(GA)** : Released two new audio-to-audio models for real-time voice\napplications using the Live API:\n  - **Gemini 3.8 Live** ([`gemini-3.8-live`](https://ai.google.dev/gemini-api/docs/models/gemini-3.8-live) ): The default option\nfor most low-latency voice agent experiences and real-time dialogue\nwithout reasoning delays. Features interleaved reasoning, default\nasynchronous function calling, and full session client content updates.\n  - **Gemini 3.8 Live Extended Thinking** ([`gemini-3.8-live-extended-thinking`](https://ai.google.dev/gemini-api/docs/models/gemini-3.8-live-extended-thinking) ): High-reasoning\naudio-to-audio model supporting background reasoning during live audio\ninteractions, recommended when higher background reasoning is required.\n To get started, see the [Live API guide](https://ai.google.dev/gemini-api/docs/live-api) , the[Capabilities guide](https://ai.google.dev/gemini-api/docs/live-api/capabilities) , and the[Thinking guide](https://ai.google.dev/gemini-api/docs/live-api/thinking) .\n\n## September 3, 2026\n\n- **Lyria 3.5 generally available (GA)** : Released the next generation of\nGoogle's music generation model:\n  - [`lyria-3.5`](https://ai.google.dev/gemini-api/docs/models/lyria-3.5) :\nFull-length song generation with improved musical coherence, natural vocals,\nand fine-grained duration and structural control.\n The model supports text and image inputs and generates high-fidelity 44.1 kHz stereo audio. See the [Music generation](https://ai.google.dev/gemini-api/docs/music-generation) guide for details and code samples.\n\n## September 2, 2026\n\n- **Gemini 3.8 Flash generally available (GA)** : Released`gemini-3.8-flash` , our most intelligent Flash model, engineered for\nlong-horizon software engineering, autonomous agents, and complex enterprise\nworkflows.To get started, see the [Gemini 3.8 Flash](https://ai.google.dev/gemini-api/docs/models/gemini-3.8-flash) model page and\nthe[Latest model guide](https://ai.google.dev/gemini-api/docs/latest-model) .\n\n## September 1, 2026\n\n- **Agentic video understanding** : Released agentic video understanding for\nGemini 3.7 Flash, 3.6 Flash, and 3.5 Flash-Lite across the Interactions and\nGenerateContent APIs. The model dynamically navigates video timelines,\nrequesting transcripts, frames, or audio tracks on demand. This approach uses\nup to 88% fewer tokens for long-form content compared to static processing.To get started, see the [Agentic video understanding](https://ai.google.dev/gemini-api/docs/video-understanding#agentic-video-understanding) guide.\n\n## August 27, 2026\n\n- **Gemini Omni Flash generally available (GA)** : Released`gemini-omni-1.1-flash` , the GA version of our fast, conversational video\ngeneration and editing model. This release includes significant new\ncapabilities:\n  - **Video extension** : Seamlessly extend existing videos by generating\ncontinuations at the end of a clip using the`extend` task or directly\nwith a prompt.\n  - **Interpolation (first + last frame)** : Generate a video transitioning\nbetween two images using the`image_to_video` task with up to 2 images.\n  - **Resolution control** : New`resolution` parameter in`video_config` supports`360p` ,`720p` (default),`1080p` , and`4k` outputs.\n1080p and 4K outputs are generated using upscaling.\n The existing `gemini-omni-flash-preview` endpoint will be deprecated on\nSeptember 30, 2026.To get started, see the [Gemini Omni Flash](https://ai.google.dev/gemini-api/docs/models/gemini-omni-flash) model page\nand the[omni guide](https://ai.google.dev/gemini-api/docs/omni) .\n\n## August 26, 2026\n\n- **Gemini 3.5 Transcribe generally available (GA)** : Released two dedicated\nspeech-to-text models based on Gemini's audio understanding:\n  - **Gemini 3.5 Transcribe** (`gemini-3.5-transcribe` ): High-accuracy,\nlow-latency non-streaming speech-to-text with utterance-based language\ndetection across 85+ languages, speaker diarization, word-level\ntimestamps, and custom vocabulary biasing (up to 1,000 terms).\n  - **Gemini 3.5 Transcribe Live** (`gemini-3.5-transcribe-live` ):\nLow-latency, bidirectional streaming speech-to-text over WebSockets using\nthe Live API, supporting interim and finalized transcription events,\nSmart transcription mode, and multiple Voice Activity Detection (VAD)\nstrategies.\n To get started, see the [Audio transcription guide](https://ai.google.dev/gemini-api/docs/transcribe) , the[Live transcription guide](https://ai.google.dev/gemini-api/docs/live-api/live-transcribe) , and the[Gemini 3.5 Transcribe model page](https://ai.google.dev/gemini-api/docs/models/gemini-3.5-transcribe) .\n\n## August 13, 2026\n\n- **Gemini 3.7 Flash generally available (GA)** : Released our most\nintelligent workhorse model yet for coding and agents:\n  - **Gemini 3.7 Flash** (`gemini-3.7-flash` ): Substantial improvements\nacross software engineering, web development, and agentic workflows,\navailable at an introductory price through December 31, 2026.\n To get started, see the [Gemini 3.7 Flash](https://ai.google.dev/gemini-api/docs/models/gemini-3.7-flash) model page\nand the[Latest model guide](https://ai.google.dev/gemini-api/docs/latest-model) .\n\n## July 30, 2026\n\n- **Gemini Robotics ER 2 in public preview** : Released two new embodied\nreasoning model endpoints for robotics:\n  - `gemini-robotics-er-2-preview` : Advanced spatial reasoning, agentic\ncode execution, multi-step tool orchestration, video moment finding,\nprogress classification, and multi-robot coordination.\n  - `gemini-robotics-er-2-streaming-preview` : Optimized for real-time\ntext streaming using the Live API, enabling low-latency robot agents with\nbidirectional audio and video input.\n Both model endpoints accept text, image, video, and audio inputs and support function calling with blocking behavior for physical robot actions. To get started, see the [Gemini Robotics ER overview](https://ai.google.dev/gemini-api/docs/robotics-overview) . For\nreal-time streaming use cases, see[Robotics with streaming](https://ai.google.dev/gemini-api/docs/robotics-streaming) .\n- **Deprecation announcement** : The`gemini-robotics-er-1.6-preview` model\nwill be[shut down](https://ai.google.dev/gemini-api/docs/deprecations) on August 31, 2026.\n\n## July 21, 2026\n\n- **Gemini 3.6 Flash and Gemini 3.5 Flash-Lite generally available (GA)** :\nReleased stable, production-ready versions of our latest 3.x Flash models:\n  - **Gemini 3.6 Flash** (`gemini-3.6-flash` ): Features improved token\nefficiency and code/agentic planning capabilities at a lower price point\nthan 3.5 Flash, resolving developer feedback around output verbosity.\n  - **Gemini 3.5 Flash-Lite** (`gemini-3.5-flash-lite` ): Offers a\nlow-latency, highly cost-effective subagent option designed for\nhigh-volume automation.\n To learn more, see the [Latest Gemini model](https://ai.google.dev/gemini-api/docs/latest-model) guide.\n- **Deprecated parameters** : The sampling parameters`temperature` ,`top_p` and`top_k` are now deprecated. See the[Latest Gemini Model](https://ai.google.dev/gemini-api/docs/latest-model#sampling-parameter-deprecation) for details.\n\n## July 6, 2026\n\n- [Developer logs](https://ai.google.dev/gemini-api/docs/logs-datasets) support for the\nInteractions API: logs for supported Interactions API calls are now viewable\nin the[AI Studio dashboard](https://aistudio.google.com/logs) .\n\n## June 30, 2026\n\n- **Gemini Omni Flash in public preview** : Released`gemini-omni-flash-preview` ,\na high-performance multimodal model designed for high-speed video generation\nand conversational video editing. Using the[Interactions API](https://ai.google.dev/gemini-api/docs/interactions-overview) ,\nyou can generate 3–10 second videos at 720p from text descriptions or animate still images,\nand then conversationally edit and refine the outputs. To get started, see the[Gemini Omni Flash](https://ai.google.dev/gemini-api/docs/omni) guide and the[Gemini Omni Flash](https://ai.google.dev/gemini-api/docs/models/gemini-omni-flash) model card.\n- Released `gemini-3.1-flash-lite-image` (Nano Banana 2 Lite) to general\navailability (GA), our built-in multimodal model optimized for ultra-low\nlatency and cost-effective image generation and editing. See the[Gemini 3.1\nFlash Lite Image](https://ai.google.dev/gemini-api/docs/models/gemini-3.1-flash-lite-image) model\ncard and the[Image generation](https://ai.google.dev/gemini-api/docs/image-generation) guide.\n\n## June 24, 2026\n\n- **Computer Use** : Launched public preview support for the[Computer Use](https://ai.google.dev/gemini-api/docs/computer-use) tool in Gemini 3.5 Flash. This\nrelease includes simplified actions with intents, built-in support for\nbrowser, mobile, and desktop environments, configurable safety policies, and\nadvanced prompt injection detection.\n\n## June 17, 2026\n\n- **Streaming support for speech generation** : Streaming via`streamGenerateContent` (and`stream: true` in the Interactions API) is now supported for the`gemini-3.1-flash-tts-preview` model. To learn more, see the[Text-to-Speech](https://ai.google.dev/gemini-api/docs/speech-generation#streaming) guide.\n\n## June 15, 2026\n\n- **Deprecation announcement** : The following image generation models are\nbeing deprecated and will be[shut down](https://ai.google.dev/gemini-api/docs/deprecations) on**August 17, 2026** :\n  - **Imagen 4 and Gemini 3 Image models** :\n    - `imagen-4.0-generate-001`\n    - `imagen-4.0-ultra-generate-001`\n    - `imagen-4.0-fast-generate-001`\n To migrate your code to newer stable or preview endpoints, refer to the [Gemini deprecations](https://ai.google.dev/gemini-api/docs/deprecations#imagen-models) page.\n- **Deprecation announcement** : The following video generation models are\nbeing deprecated and will be[shut down](https://ai.google.dev/gemini-api/docs/deprecations) on**June 30, 2026** :\n  - **Veo models** :\n    - `veo-2.0-generate-001`\n    - `veo-3.0-generate-001`\n    - `veo-3.0-fast-generate-001`\n Update your integration to either use the Veo 3.1 preview model IDs ( `veo-3.1-generate-preview` ,`veo-3.1-fast-generate-preview` ) or the\n3.1 GA models available through the[Gemini Enterprise Agent Platform](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/veo/3-1-generate) to avoid service interruptions.\n- **Deprecation announcement** : The experimental GMP Contextual View tool (a fixed interface for Grounding with Google Maps outputs) will[shut down](https://ai.google.dev/gemini-api/docs/deprecations) on**June 15, 2026** :\n\n## June 1, 2026\n\n- The following Gemini 2.0 models are now [shut down](https://ai.google.dev/gemini-api/docs/deprecations) :\n  - `gemini-2.0-flash`\n  - `gemini-2.0-flash-001`\n  - `gemini-2.0-flash-lite`\n  - `gemini-2.0-flash-lite-001`\n Use [`gemini-3.5-flash`](https://ai.google.dev/gemini-api/docs/models/gemini-3.5-flash) or[` gemini-3.1-flash-lite`](https://ai.google.dev/gemini-api/docs/models/gemini-3.1-flash-lite) instead.\n\n## May 28, 2026\n\n- Released `gemini-3.1-flash-image` (Nano Banana 2) and`gemini-3-pro-image` (Nano Banana Pro), the generally available (GA) versions of our native\nvisual models,[Gemini 3.1 Flash Image](https://ai.google.dev/gemini-api/docs/models/gemini-3.1-flash-image) and[Gemini 3 Pro Image](https://ai.google.dev/gemini-api/docs/models/gemini-3-pro-image) .\n- **Video-to-image generation support** : You can now pass a video file (via\ndirect upload or as a public YouTube URL) as multimodal context alongside a\ntext prompt to generate high-quality thumbnails, cinematic movie posters, or\nsummary infographics. This feature is supported exclusively on the`gemini-3.1-flash-image` model. To learn more, see the[Video-to-image generation](https://ai.google.dev/gemini-api/docs/image-generation#video-to-image) guide.\n- Deprecation announcement: The `gemini-3.1-flash-image-preview` and`gemini-3-pro-image-preview` models are deprecated\nand will be[shut down](https://ai.google.dev/gemini-api/docs/deprecations) on June 25, 2026.\n\n## May 25, 2026\n\n- The `gemini-3.1-flash-lite-preview` model has been[shut down](https://ai.google.dev/gemini-api/docs/deprecations) . Use[`gemini-3.1-flash-lite`](https://ai.google.dev/gemini-api/docs/models/gemini-3.1-flash-lite) instead.\n\n## May 19, 2026\n\n- Released `gemini-3.5-flash` , the generally available (GA) version of[Gemini 3.5 Flash](https://ai.google.dev/gemini-api/docs/models/gemini-3.5-flash) ,\nour most intelligent model for sustained frontier performance on\nagentic and coding tasks. This is now the model behind`gemini-flash-latest` .\n- Launched the **Managed Agents in the Gemini API** in public preview. This enables\ndevelopers to build and deploy autonomous, stateful agents that run in\nsecure, isolated Google-hosted Linux sandbox environments. To learn more,\nsee the[Agents overview](https://ai.google.dev/gemini-api/docs/agents) page and the[Quickstart](https://ai.google.dev/gemini-api/docs/managed-agents-quickstart) .\n- Released the general-purpose **Antigravity Agent** managed agent,[`antigravity-preview-05-2026`](https://ai.google.dev/gemini-api/docs/models/antigravity-preview-05-2026) , in public preview.\nThe Antigravity agent can autonomously plan, reason, write and execute code,\nmanage files, and browse the web inside its sandbox container. See the[Antigravity Agent](https://ai.google.dev/gemini-api/docs/antigravity-agent) guide for code\nsamples and specifications.\n\n## May 7, 2026\n\n- Released `gemini-3.1-flash-lite` , the generally available (GA) version of[Gemini 3.1 Flash-Lite](https://ai.google.dev/gemini-api/docs/models/gemini-3.1-flash-lite) ,\noptimized for speed, scale, and cost efficiency.\n- Deprecation announcement: The `gemini-3.1-flash-lite-preview` model is\ndeprecating on 5/11/26 and will be[shut down](https://ai.google.dev/gemini-api/docs/deprecations) on May 25, 2026.\n\n## May 6, 2026\n\n- **Upcoming breaking change** : The[Interactions API](https://ai.google.dev/gemini-api/docs/interactions-overview) request and response schema (`outputs` →`steps` ) and output format\nconfiguration (`response_format` ) are changing. The new schema becomes the\ndefault on**May 26** and the legacy schema will be removed on**June 8** .\nSee the[migration guide](https://ai.google.dev/gemini-api/docs/interactions-breaking-changes-may-2026) for details.\n\n## May 5, 2026\n\n- Updated **File Search** to support multimodal search. You can now natively\nembed and search through images using the`gemini-embedding-2` model.\nGrounding metadata now includes`media_id` for visual citations and`page_numbers` that indicate where information is found. To learn\nmore, see the[File Search](https://ai.google.dev/gemini-api/docs/file-search) guide.\n\n## May 4, 2026\n\n- Launched event-driven [Webhooks](https://ai.google.dev/gemini-api/docs/webhooks) support in the\nGemini API to replace polling workflows for the Batch API and long-running\noperations.\n\n## April 30, 2026\n\n- The `gemini-robotics-er-1.5-preview` model has been[shut down](https://ai.google.dev/gemini-api/docs/deprecations) . Use[`gemini-robotics-er-1.6-preview`](https://ai.google.dev/gemini-api/docs/models/gemini-robotics-er-1.6-preview) instead.\n\n## April 22, 2026\n\n- Released `gemini-embedding-2` as generally available\n(GA). To learn more, see the[Embeddings](https://ai.google.dev/gemini-api/docs/embeddings) page.\n\n## April 21, 2026\n\n- Released new versions of the [Deep Research](https://ai.google.dev/gemini-api/docs/deep-research) agent with collaborative planning, visualization support, MCP server\nintegration, and File Search:\n  - [`deep-research-preview-04-2026`](https://ai.google.dev/gemini-api/docs/models/deep-research-preview-04-2026) : Designed for\nspeed and efficiency, ideal to be streamed back to a client UI.\n  - [`deep-research-max-preview-04-2026`](https://ai.google.dev/gemini-api/docs/models/deep-research-max-preview-04-2026) : Maximum\ncomprehensiveness for automated context gathering and synthesis.\n\n## April 15, 2026\n\n- Launched [Gemini 3.1 Flash TTS Preview](https://ai.google.dev/gemini-api/docs/models/gemini-3.1-flash-tts-preview) , our cost-efficient,\nexpressive, and steerable text to speech model. Read the[Text-to-Speech](https://ai.google.dev/gemini-api/docs/speech-generation) docs to learn more.\n\n## April 14, 2026\n\n- Released `gemini-robotics-er-1.6-preview` , our updated robotics model.\nIt now has new capabilities like instrument reading, improved spatial and\nphysical reasoning capabilities. To learn more, see[Gemini Robotics ER](https://ai.google.dev/gemini-api/docs/robotics-overview) page and the[blog](https://deepmind.google/blog/gemini-robotics-er-1-6) .\n- Deprecation announcement: The `gemini-robotics-er-1.5-preview` model\nwill be[shut down](https://ai.google.dev/gemini-api/docs/deprecations) on April 30, 2026 at 9AM\nPST.\n\n## April 2, 2026\n\n- Released `gemma-4-26b-a4b-it` and`gemma-4-31b-it` , available on[AI Studio](https://aistudio.google.com) and through the Gemini API,\nas part of the[Gemma 4](https://ai.google.dev/gemma/docs/core) launch.\n\n## April 1, 2026\n\n- Introduced the new [Flex](https://ai.google.dev/gemini-api/docs/flex-inference) and[Priority](https://ai.google.dev/gemini-api/docs/priority-inference) inference tiers, offering more options\nfor optimizing cost or latency.\n\n## March 31, 2026\n\n- Launched Veo 3.1 Lite Preview, [`veo-3.1-lite-generate-preview`](https://ai.google.dev/gemini-api/docs/models/veo-3.1-lite-generate-preview) , our most\ncost-efficient[video generation](https://ai.google.dev/gemini-api/docs/video) model, designed\nfor rapid iteration and building high-volume applications.\n- The `gemini-2.5-flash-lite-preview-09-2025` model has been shut down. Use[`gemini-3.1-flash-lite-preview`](https://ai.google.dev/gemini-api/docs/models/gemini-3.1-flash-lite-preview) instead.\n\n## March 26, 2026\n\n- Released [`gemini-3.1-flash-live-preview`](https://ai.google.dev/gemini-api/docs/models/gemini-3.1-flash-live-preview) , the latest\naudio-to-audio (A2A) model designed for real-time dialogue and voice-first\nAI applications. Read the[Live API](https://ai.google.dev/gemini-api/docs/live-api) docs to get\nstarted.\n\n## March 25, 2026\n\n- Launched [Lyria 3](https://ai.google.dev/gemini-api/docs/music-generation) music generation\nmodels:[`lyria-3-clip-preview`](https://ai.google.dev/gemini-api/docs/models/lyria-3-clip-preview) (30-second clips) and[`lyria-3-pro-preview`](https://ai.google.dev/gemini-api/docs/models/lyria-3-pro-preview) (full-length songs). Both models accept text and image inputs and generate\nhigh-quality, 48kHz stereo audio. See the[Music generation](https://ai.google.dev/gemini-api/docs/music-generation) guide for details and\ncode samples.\n\n## March 23, 2026\n\n- Rolled out [Prepay and Postpay billing plans](https://ai.google.dev/gemini-api/docs/billing) in\nAI Studio. Existing accounts may be affected; read the[Billing](https://ai.google.dev/gemini-api/docs/billing) documentation for more information.\n\n## March 18, 2026\n\n- Released the new [Built-in Tools and Function Calling Combination](https://ai.google.dev/gemini-api/docs/tool-combination) feature, making it possible\nto use Gemini's built-in tools alongside custom function calling\ntools in a single API call.\n- [Grounding with Google Maps](https://ai.google.dev/gemini-api/docs/maps-grounding#supported_models) is now supported for Gemini 3 models going forward.\n\n## March 16, 2026\n\n- Introduced revamped [Usage Tiers](https://ai.google.dev/gemini-api/docs/billing#about-billing) and[Billing Account spend caps](https://ai.google.dev/gemini-api/docs/billing#tier-spend-caps) for a better user billing experience.\n\n## March 12, 2026\n\n- Introduced [project-level spend caps](https://ai.google.dev/gemini-api/docs/billing#project-spend-caps) to billing in AI Studio.\n\n## March 10, 2026\n\n- Released `gemini-embedding-2-preview` , our first multimodal embedding model.\nIt supports text, image, video, audio, and PDF inputs,\nmapping all modalities into a unified embedding space. To learn more, see[Embeddings](https://ai.google.dev/gemini-api/docs/embeddings) .\n- Deprecation announcement: The `gemini-2.5-flash-lite-preview-09-2025` model\nwill be[shut down](https://ai.google.dev/gemini-api/docs/deprecations) on March 31, 2026.\n\n## March 9, 2026\n\n- The Gemini 3 Pro Preview model has been [shut down](https://ai.google.dev/gemini-api/docs/deprecations) . The`gemini-3-pro-preview` now points to[`gemini-3.1-pro-preview`](https://ai.google.dev/gemini-api/docs/models/gemini-3.1-pro-preview) .\n\n## March 3, 2026\n\n- Launched Gemini 3.1 Flash-Lite Preview, the first Flash-Lite model in the\nGemini 3 series. Read the [model page](https://ai.google.dev/gemini-api/docs/models/gemini-3.1-flash-lite-preview) for specs, specific\nupdates, and developer guidance.\n\n## February 26, 2026\n\n- Launched Nano Banana 2, [Gemini 3.1 Flash Image Preview](https://ai.google.dev/gemini-api/docs/models/gemini-3.1-flash-image-preview) , a high-efficiency\nmodel optimized for speed and high-volume use cases.\n- Deprecation announcement: Gemini 3 Pro Preview ( `gemini-3-pro-preview` )\nwill be[shut down](https://ai.google.dev/gemini-api/docs/deprecations) March 9, 2026.\n\n## February 19, 2026\n\n- Released [Gemini 3.1 Pro Preview](https://ai.google.dev/gemini-api/docs/models/gemini-3.1-pro-preview) , our latest iteration in\nthe new Gemini 3 series family.\n- Launched a separate endpoint `gemini-3.1-pro-preview-customtools` , which is\nbetter at prioritizing custom tools, for users building with a mix of bash\nand tools.\n\n## February 18, 2026\n\n- Deprecation announcement: The following models will be [shut down](https://ai.google.dev/gemini-api/docs/deprecations) June 1, 2026:\n  - `gemini-2.0-flash`\n  - `gemini-2.0-flash-001`\n  - `gemini-2.0-flash-lite`\n  - `gemini-2.0-flash-lite-001`\n\n## February 17, 2026\n\n- The following models are [shut down](https://ai.google.dev/gemini-api/docs/deprecations) :\n  - `gemini-2.5-flash-preview-09-25`\n  - `imagen-4.0-generate-preview-06-06`\n  - `imagen-4.0-ultra-generate-preview-06-06`\n\n## January 29, 2026\n\n- Launched support for the Computer Use tool in `gemini-3-pro-preview` and`gemini-3-flash-preview` .\n\n## January 21, 2026\n\n- Changed the `latest` aliases:\n  - `gemini-pro-latest` switched to`gemini-3-pro-preview`\n  - `gemini-flash-latest` switched to`gemini-3-flash-preview`\n\n## January 15, 2026\n\n- Deprecation announcement: The following models will be [shut down](https://ai.google.dev/gemini-api/docs/deprecations) February 17, 2026:\n  - `gemini-2.5-flash-preview-09-25`\n  - `imagen-4.0-generate-preview-06-06`\n  - `imagen-4.0-ultra-generate-preview-06-06`\n- The `gemini-2.5-flash-image-preview` model has been shut down.\n\n## January 14, 2026\n\n- The `text-embedding-004` model has been[shut down](https://ai.google.dev/gemini-api/docs/deprecations) .\n\n## January 13, 2026\n\n- Added 4k output resolutions for [Veo](https://ai.google.dev/gemini-api/docs/video) and more\nsupport for portrait videos in all resolutions.\n\n## January 12, 2026\n\n- Launched model lifecycle feature. Some models will now specify the lifecycle stage and deprecation timeline. See the following documentation for more information:\n\n## January 8, 2026\n\n- Launched support for Cloud Storage buckets and any public and private DB\npre-signed URL as data input source for the Gemini API. The file size limit\nhas also increased from 20MB to 100MB. For details, see [File input methods\nguide](https://ai.google.dev/gemini-api/docs/file-input-methods) .\n\n## December 19, 2025\n\n- Introduced a breaking change to the Interactions API in\nv1beta. The `total_reasoning_tokens` field has been renamed to`total_thought_tokens` to better align with the concept of \"thoughts\" in\nthinking models.\n\n## December 17, 2025\n\n- Launched Gemini 3 Flash Preview, `gemini-3-flash-preview` , delivering fast\nfrontier-class performance that rivals larger models at a fraction of the\ncost. With upgraded visual and spatial reasoning, and agentic coding\ncapabilities. Read the documentation on some new features, including:\n\n## December 12, 2025\n\n- Released `gemini-2.5-flash-native-audio-preview-12-2025` ,\na new native audio model for the Live API. This update improves the model's\nability to handle complex workflows. To learn more, see the[Live API guide](https://ai.google.dev/gemini-api/docs/live-guide) and[Gemini 2.5 Flash Native Audio](https://ai.google.dev/gemini-api/docs/models/gemini-2.5-flash-live) .\n\n## December 11, 2025\n\n- Launched the Interactions API. This API provides a unified interface for interacting with Gemini models and agents. To learn more, see the [Interactions API](https://ai.google.dev/gemini-api/docs/interactions-overview) guide.\n- Launched the Gemini Deep Research agent in preview. It can autonomously plan, execute, and synthesize results for multi-step research tasks. See the [Deep Research](https://ai.google.dev/gemini-api/docs/deep-research) guide for\ndetails.\n\n## December 10, 2025\n\n- Launched enhancements to our [text-to-speech models](https://ai.google.dev/gemini-api/docs/speech-generation) , Gemini 2.5 Flash TTS preview\n(optimized for low latency) and Gemini 2.5 Pro TTS preview (optimized for\nquality), including enhanced expressivity, precision pacing, and seamless\ndialogue.\n\n## December 9, 2025\n\n- The following Gemini Live API models are now shut down:\n  - `gemini-2.0-flash-live-001`\n  - `gemini-live-2.5-flash-preview`\n\n## December 5, 2025\n\n- Gemini 3 billing for [Grounding with Google Search](https://ai.google.dev/gemini-api/docs/google-search) will begin on January 5, 2026.\n\n## December 4, 2025\n\n- Deprecation announcement: The `gemini-2.5-flash-image-preview` model will be\nshut down January 15, 2026.\n\n## December 3, 2025\n\n- Deprecation announcement: The `text-embedding-004` model will be shut down\nJanuary 14, 2026.\n\n## November 20, 2025\n\n- Released Gemini 3 Pro Image Preview, `gemini-3-pro-image-preview` , the\nnext iteration to the Nano Banana model. Read the[Image generation](https://ai.google.dev/gemini-api/docs/image-generation) page for more details.\n\n## November 18, 2025\n\n- Launched the first Gemini 3 series model, `gemini-3-pro-preview` , our\nstate-of-the-art reasoning and multimodal understanding model with powerful\nagentic and coding capabilities.In addition to improvements in intelligence and performance, Gemini 3 Pro Preview introduces new behavior around: Read the [Gemini 3 Developer Guide](https://ai.google.dev/gemini-api/docs/gemini-3) for\nmigration, new features, and specs.\n\n## November 11, 2025\n\n- Deprecation announcement: The following models will be shut down: \n  - November 12: \n    - `veo-3.0-fast-generate-preview`\n    - `veo-3.0-generate-preview`\n  - November 14: \n    - `gemini-2.0-flash-exp-image-generation`\n    - `gemini-2.0-flash-preview-image-generation`\n\n## November 10, 2025\n\n- The following model is shut down: \n  - `imagen-3.0-generate-002`\n Use [Imagen 4](https://ai.google.dev/gemini-api/docs/imagen#imagen-4) instead. Refer to the[Gemini deprecations table](https://ai.google.dev/gemini-api/docs/deprecations) for more details.\n\n## November 6, 2025\n\n- Launched the File Search API to public preview, enabling developers to\nground responses in their own data. Read the new [File Search](https://ai.google.dev/gemini-api/docs/file-search) page for more info.\n\n## November 4, 2025\n\n- For [Gemini 2.5 Flash Image](https://ai.google.dev/gemini-api/docs/image-generation) , the input\ntoken count for images has been reduced from 1290 to 258, lowering the cost\nof image editing.\n- Deprecation announcement: The following models will be shut down: \n  - November 18th: \n    - `gemini-2.5-flash-lite-preview-06-17`\n    - `gemini-2.5-flash-preview-05-20`\n  - December 2nd: \n    - `gemini-2.0-flash-thinking-exp`\n    - `gemini-2.0-flash-thinking-exp-01-21`\n    - `gemini-2.0-flash-thinking-exp-1219`\n    - `gemini-2.5-pro-preview-03-25`\n    - `gemini-2.5-pro-preview-05-06`\n    - `gemini-2.5-pro-preview-06-05`\n  - December 9th: \n    - `gemini-2.0-flash-lite-preview`\n    - `gemini-2.0-flash-lite-preview-02-05`\n    - `gemini-2.0-flash-exp`\n    - `gemini-2.0-pro-exp`\n    - `gemini-2.0-pro-exp-02-05`\n\n## October 29, 2025\n\n- Launched the new [logging and datasets](https://ai.google.dev/gemini-api/docs/logs-datasets) tool\nfor the Gemini API.\n\n## October 20, 2025\n\n- The following Gemini Live API models are now shut down: \n  - `gemini-2.5-flash-preview-native-audio-dialog`\n  - `gemini-2.5-flash-exp-native-audio-thinking-dialog`\n You can use `gemini-2.5-flash-native-audio-preview-09-2025` instead.\n- Deprecation announcement: Shut down for `gemini-2.0-flash-live-001` and`gemini-live-2.5-flash-preview` coming December 09, 2025.\n\n## October 17, 2025\n\n- **Grounding with Google Maps** is now\ngenerally available. For more information, see[Grounding with Google Maps](https://ai.google.dev/gemini-api/docs/maps-grounding) documentation.\n\n## October 15, 2025\n\n- Released [Veo 3.1 and 3.1 Fast](https://ai.google.dev/gemini-api/docs/video#veo-3.1) models in\npublic preview, with new features including:\n  - Extending Veo-created videos.\n  - Referencing up to three images to generate a video.\n  - Providing first and last frame images to generate videos from.\n This launch also added more options for Veo 3 output video durations: 4, 6, and 8 seconds.\n- Deprecation announcement: Shut down for `veo-3.0-generate-preview` and`veo-3.0-fast-generate-preview` coming November 12, 2025.\n\n## October 7, 2025\n\n- Launched [Gemini 2.5 Computer Use Preview](https://ai.google.dev/gemini-api/docs/computer-use)\n\n## October 2, 2025\n\n- Launched Gemini 2.5 Flash Image GA: [Image Generation with Gemini](https://ai.google.dev/gemini-api/docs/image-generation)\n\n## September 29, 2025\n\n- The following Gemini 1.5 models are now shut down:\n  - `gemini-1.5-pro`\n  - `gemini-1.5-flash-8b`\n  - `gemini-1.5-flash`\n\n## September 25, 2025\n\n- Released Gemini Robotics ER 1.5 model in preview. See the [Robotics overview](https://ai.google.dev/gemini-api/docs/robotics-overview) to learn about how to use the model for your robotics application.\n- Launched following preview models: \n  - `gemini-2.5-flash-preview-09-2025`\n  - `gemini-2.5-flash-lite-preview-09-2025`\n See the [Models](https://ai.google.dev/gemini-api/docs/models) page for details.\n\n## September 23, 2025\n\n- Released `gemini-2.5-flash-native-audio-preview-09-2025` ,\na new native audio model for the Live API with improved function calling\nand speech cut off handling. To learn more, see the[Live API guide](https://ai.google.dev/gemini-api/docs/live-guide) and[Gemini 2.5 Flash Native Audio](https://ai.google.dev/gemini-api/docs/models#gemini-2.5-flash-native-audio) .\n\n## September 16, 2025\n\n- Deprecation announcement: The following models will be shut down in October 2025: \n  - `embedding-001`\n  - `embedding-gecko-001`\n  - `gemini-embedding-exp-03-07` (`gemini-embedding-exp` )\n See the [Embeddings](https://ai.google.dev/gemini-api/docs/embeddings) page for details on the latest embeddings\nmodel.\n\n## September 10, 2025\n\n- Released support for the\n[Embeddings model in Batch API](https://ai.google.dev/gemini-api/docs/batch-api#batch-embedding) ,\nand added Batch API to the[OpenAI compatibility library](https://ai.google.dev/gemini-api/docs/openai#batch) for even\neasier ways to get started with batch queries.\n\n## September 9, 2025\n\n- Launched Veo 3 and Veo 3 Fast GA, with lower pricing and new options for\naspect ratios, resolution, and seeding. Read the\n[Veo documentation](https://ai.google.dev/gemini-api/docs/video#model-features) for more\ninformation.\n\n## August 26, 2025\n\n- Launched [Gemini 2.5 Image Preview](https://ai.google.dev/gemini-api/docs/models#gemini-2.5-flash-image-preview) ,\nour latest native image generation model.\n\n## August 18, 2025\n\n- Released [URL context tool](https://ai.google.dev/gemini-api/docs/url-context) to general\navailability (GA), a tool for providing URLs as additional context to\nprompts. Support for using URL context with the`gemini-2.0-flash` model\n(available during experimental release) will be discontinued in one week.\n\n## August 14, 2025\n\n- Released Imagen 4 Ultra, Standard and Fast models as generally available\n(GA). To learn more, see the [Imagen](https://ai.google.dev/gemini-api/docs/imagen) page.\n\n## August 7, 2025\n\n- `allow_adult` setting in Image to Video generation are now available in\nrestricted regions. See the[Veo](https://ai.google.dev/gemini-api/docs/video?example=dialogue#veo-model-parameters) page for details.\n\n## July 31, 2025\n\n- Launched image-to-video generation for the Veo 3 Preview model.\n- Released Veo 3 Fast Preview model.\n- To learn more about Veo 3, visit the [Veo](https://ai.google.dev/gemini-api/docs/video) page.\n\n## July 22, 2025\n\n- Released `gemini-2.5-flash-lite` , our fast, low-cost, high-performance Gemini\n2.5 model. To learn more, see[Gemini 2.5\nFlash-Lite](https://ai.google.dev/gemini-api/docs/models#gemini-2.5-flash-lite) .\n\n## July 17, 2025\n\n- Launched `veo-3.0-generate-preview` , the latest update to Veo introducing\nvideo with audio generation. To learn more about Veo 3, visit the[Veo](https://ai.google.dev/gemini-api/docs/video) page.\n- Increased rate limits for Imagen 4 Standard and Ultra. Visit the [Rate limits](https://ai.google.dev/gemini-api/docs/rate-limits) page for more details.\n\n## July 14, 2025\n\n- Released `gemini-embedding-001` , the stable version of our\ntext embedding model. To learn more, see[embeddings](https://ai.google.dev/gemini-api/docs/embeddings) . The`gemini-embedding-exp-03-07` model will be deprecated on August 14, 2025.\n\n## July 7, 2025\n\n- Launched Gemini API Batch Mode. Batch up requests and send them to process\nasynchronously. To learn more, see [Batch Mode](https://ai.google.dev/gemini-api/docs/batch-mode) .\n\n## June 26, 2025\n\n- The preview models `gemini-2.5-pro-preview-05-06` and`gemini-2.5-pro-preview-03-25` are now redirecting to\nthe latest stable version`gemini-2.5-pro` .\n- `gemini-2.5-pro-exp-03-25` is shut down.\n\n## June 24, 2025\n\n- Released Imagen 4 Ultra and Standard Preview models. To learn more, see the\n[Image generation](https://ai.google.dev/gemini-api/docs/image-generation) page.\n\n## June 17, 2025\n\n- Released `gemini-2.5-pro` , the stable version of our most powerful\nmodel, now with adaptive thinking. To learn more, see[Gemini 2.5 Pro](https://ai.google.dev/gemini-api/docs/models#gemini-2.5-pro) and[Thinking](https://ai.google.dev/gemini-api/docs/thinking) .`gemini-2.5-pro-preview-05-06` will be redirected to`gemini-2.5-pro` on June 26, 2025.\n- Released `gemini-2.5-flash` , our first stable 2.5 Flash model. To learn\nmore, see[Gemini 2.5 Flash](https://ai.google.dev/gemini-api/docs/models#gemini-2.5-flash) .`gemini-2.5-flash-preview-04-17` will be deprecated on July 15, 2025.\n- Released `gemini-2.5-flash-lite-preview-06-17` , a low-cost, high-performance\nGemini 2.5 model. To learn more, see[Gemini 2.5 Flash-Lite\nPreview](https://ai.google.dev/gemini-api/docs/models#gemini-2.5-flash-lite) .\n\n## June 05, 2025\n\n- Released `gemini-2.5-pro-preview-06-05` , a new version of our most powerful\nmodel, now with adaptive thinking. To learn more, see[Gemini 2.5 Pro Preview](https://ai.google.dev/gemini-api/docs/models#gemini-2.5-pro-preview-06-05) and[Thinking](https://ai.google.dev/gemini-api/docs/thinking) .`gemini-2.5-pro-preview-05-06` will be redirected to`gemini-2.5-pro` on\nJune 26, 2025.\n\n## May 27, 2025\n\n- The last available tuning model, Gemini 1.5 Flash 001, has been shut down.\nTuning is no longer supported on any models.\nSee [Fine tuning with the Gemini API](https://ai.google.dev/gemini-api/docs/model-tuning) .\n\n## May 20, 2025\n\n**API updates:**\n\n- Launched support for\n[custom video preprocessing](https://ai.google.dev/gemini-api/docs/video-understanding#customize-video-processing) using clipping intervals and configurable frame rate sampling.\n- Launched multi-tool use, which supports configuring\n[code execution](https://ai.google.dev/gemini-api/docs/code-execution) and[Grounding with Google Search](https://ai.google.dev/gemini-api/docs/grounding) on the same`generateContent` request.\n- Launched support for\n[asynchronous function calls](https://ai.google.dev/gemini-api/docs/live-tools#async-function-calling) in the Live API.\n- Launched an experimental\n[URL context tool](https://ai.google.dev/gemini-api/docs/url-context) for providing URLs as additional context to prompts.\n\n**Model updates:**\n\n- Released `gemini-2.5-flash-preview-05-20` , a Gemini[preview](https://ai.google.dev/gemini-api/docs/models#model-versions) model optimized for\nprice-performance and adaptive thinking. To learn more, see[Gemini 2.5 Flash Preview](https://ai.google.dev/gemini-api/docs/models#gemini-2.5-flash-preview) and[Thinking](https://ai.google.dev/gemini-api/docs/thinking) .\n- Released the\n[`gemini-2.5-pro-preview-tts`](https://ai.google.dev/gemini-api/docs/models#gemini-2.5-pro-preview-tts) and[` gemini-2.5-flash-preview-tts`](https://ai.google.dev/gemini-api/docs/models#gemini-2.5-flash-preview-tts) models, which are capable of[generating speech](https://ai.google.dev/gemini-api/docs/speech-generation) with one or two\nspeakers.\n- Released the `lyria-realtime-exp` model, which[generates music](https://ai.google.dev/gemini-api/docs/music-generation) in real time.\n- Released `gemini-2.5-flash-preview-native-audio-dialog` and`gemini-2.5-flash-exp-native-audio-thinking-dialog` ,\nnew Gemini models for the Live API with native audio output capabilities. To\nlearn more, see the[Live API guide](https://ai.google.dev/gemini-api/docs/live-guide#native-audio-output) and[Gemini 2.5 Flash Native Audio](https://ai.google.dev/gemini-api/docs/models#gemini-2.5-flash-native-audio) .\n- Released `gemma-3n-e4b-it` preview, available on[AI Studio](https://aistudio.google.com) and through the Gemini API,\nas part of the[Gemma 3n](https://ai.google.dev/gemma/docs/3n) launch.\n\n## May 7, 2025\n\n- Released `gemini-2.0-flash-preview-image-generation` , a preview model for\ngenerating and editing images. To learn more, see[Image\ngeneration](https://ai.google.dev/gemini-api/docs/image-generation) and[Gemini 2.0 Flash Preview Image\nGeneration](https://ai.google.dev/gemini-api/docs/models#gemini-2.0-flash-preview-image-generation) .\n\n## May 6, 2025\n\n- Released `gemini-2.5-pro-preview-05-06` , a new version of our most powerful\nmodel, with improvements on code and function calling.`gemini-2.5-pro-preview-03-25` will automatically point to the new version of the model.\n\n## April 17, 2025\n\n- Released `gemini-2.5-flash-preview-04-17` , a Gemini[preview](https://ai.google.dev/gemini-api/docs/models#model-versions) model optimized for\nprice-performance and adaptive thinking. To learn more, see[Gemini 2.5 Flash Preview](https://ai.google.dev/gemini-api/docs/models#gemini-2.5-flash-preview) and[Thinking](https://ai.google.dev/gemini-api/docs/thinking) .\n\n## April 16, 2025\n\n- Launched context caching for\n[Gemini 2.0 Flash](https://ai.google.dev/gemini-api/docs/models#gemini-2.0-flash) .\n\n## April 9, 2025\n\n**Model updates:**\n\n- Released `veo-2.0-generate-001` , a generally available (GA) text- and\nimage-to-video model, capable of generating detailed and artistically\nnuanced videos. To learn more, see the[Veo docs](https://ai.google.dev/gemini-api/docs/video) .\n- Released `gemini-2.0-flash-live-001` , a public preview version of the[Live API](https://ai.google.dev/gemini-api/docs/live) model with billing enabled.\n  - **Enhanced Session Management and Reliability**\n    - **Session Resumption:** Keep sessions alive across temporary network\ndisruptions. The API now supports server-side session state storage (for\nup to 24 hours) and provides handles (session_resumption) to reconnect\nand resume where you left off.\n    - **Longer Sessions via Context Compression:** Enable extended\ninteractions beyond previous time limits. Configure context window\ncompression with a sliding window mechanism to automatically manage\ncontext length, preventing abrupt terminations due to context limits.\n    - **Graceful Disconnect Notification:** Receive a`GoAway` server\nmessage indicating when a connection is about to close, allowing for\ngraceful handling before termination.\n  - **More Control over Interaction Dynamics**\n  - **Configurable Voice Activity Detection (VAD):** Choose sensitivity\nlevels or disable automatic VAD entirely and use new client events\n(`activityStart` ,`activityEnd` ) for manual turn control.\n  - **Configurable Interruption Handling:** Decide whether user input\nshould interrupt the model's response.\n  - **Configurable Turn Coverage:** Choose whether the API processes all\naudio and video input continuously or only captures it when the end-user\nis detected speaking.\n  - **Configurable Media Resolution:** Optimize for quality or token usage\nby selecting the resolution for input media.\n  - **Richer Output and Features**\n  - **Expanded Voice & Language Options:** Choose from two new voices and\n30 new languages for audio output. The output language is now\nconfigurable within`speechConfig` .\n  - **Text Streaming:** Receive text responses incrementally as they are\ngenerated, enabling faster display to the user.\n  - **Token Usage Reporting:** Gain insights into usage with detailed\ntoken counts provided in the`usageMetadata` field of server messages,\nbroken down by modality and prompt or response phases.\n\n## April 4, 2025\n\n- Released `gemini-2.5-pro-preview-03-25` , a public preview Gemini 2.5 Pro version\nwith billing enabled. You can continue to use`gemini-2.5-pro-exp-03-25` on\nthe free tier.\n\n## March 25, 2025\n\n- Released `gemini-2.5-pro-exp-03-25` , a public experimental Gemini model\nwith thinking mode always on by default.\nTo learn more, see[Gemini 2.5 Pro Experimental](https://ai.google.dev/gemini-api/docs/models#gemini-2.5-pro-preview-03-25) .\n\n## March 12, 2025\n\n**Model updates:**\n\n- Launched an experimental [Gemini 2.0 Flash](https://ai.google.dev/gemini-api/docs/image-generation#gemini) model capable of image generation and editing.\n- Released `gemma-3-27b-it` , available on[AI Studio](https://aistudio.google.com) and through the Gemini API,\nas part of the[Gemma 3](https://ai.google.dev/gemma/docs/core) launch.\n\n**API updates:**\n\n- Added support for\n[YouTube URLs](https://ai.google.dev/gemini-api/docs/vision#youtube) as a media source.\n- Added support for including an\n[inline video](https://ai.google.dev/gemini-api/docs/vision#inline-video) of less than 20MB.\n\n## March 11, 2025\n\n**SDK updates:**\n\n- Released the\n[Google Gen AI SDK for TypeScript and JavaScript](https://googleapis.github.io/js-genai) to public preview.\n\n## March 7, 2025\n\n**Model updates:**\n\n- Released `gemini-embedding-exp-03-07` , an[experimental](https://ai.google.dev/gemini-api/docs/models/experimental-models) Gemini-based embeddings model in public preview.\n\n## February 28, 2025\n\n**API updates:**\n\n- Support for [Search as a tool](https://ai.google.dev/gemini-api/docs/grounding) added to`gemini-2.0-pro-exp-02-05` , an experimental model based on\nGemini 2.0 Pro.\n\n## February 25, 2025\n\n**Model updates:**\n\n- Released `gemini-2.0-flash-lite` , a generally available (GA) version of[Gemini 2.0 Flash-Lite](https://ai.google.dev/gemini-api/docs/models/gemini#gemini-2.0-flash-lite) ,\nwhich is optimized for speed, scale, and cost efficiency.\n\n## February 19, 2025\n\n**AI Studio updates:**\n\n- Support for\n[additional regions](https://ai.google.dev/gemini-api/docs/available-regions) (Kosovo, Greenland and Faroe Islands).\n\n**API updates:**\n\n- Support for\n[additional regions](https://ai.google.dev/gemini-api/docs/available-regions) (Kosovo, Greenland and Faroe Islands).\n\n## February 18, 2025\n\n**Model updates:**\n\n- Gemini 1.0 Pro is no longer supported. For the list of supported models, see\n[Gemini models](https://ai.google.dev/gemini-api/docs/models/gemini) .\n\n## February 11, 2025\n\n**API updates:**\n\n- Updates on the\n[OpenAI libraries compatibility](https://ai.google.dev/gemini-api/docs/openai) .\n\n## February 6, 2025\n\n**Model updates:**\n\n- Released `imagen-3.0-generate-002` , a generally available (GA) version of[Imagen 3 in the Gemini API](https://ai.google.dev/gemini-api/docs/imagen) .\n\n**SDK updates:**\n\n- Released the [Google Gen AI SDK for Java](https://github.com/googleapis/java-genai) for public preview.\n\n## February 5, 2025\n\n**Model updates:**\n\n- Released `gemini-2.0-flash-001` , a generally available (GA) version of[Gemini 2.0 Flash](https://ai.google.dev/gemini-api/docs/models/gemini#gemini-2.0-flash) that\nsupports text-only output.\n- Released `gemini-2.0-pro-exp-02-05` ,\nan[experimental](https://ai.google.dev/gemini-api/docs/models/experimental-models) public\npreview version of Gemini 2.0 Pro.\n- Released `gemini-2.0-flash-lite-preview-02-05` , an experimental public\npreview[model](https://ai.google.dev/gemini-api/docs/models/gemini#gemini-2.0-flash-lite) optimized for cost efficiency.\n\n**API updates:**\n\n- Added\n[file input and graph output](https://ai.google.dev/gemini-api/docs/code-execution#input-output) support to code execution.\n\n**SDK updates:**\n\n- Released the\n[Google Gen AI SDK for Python](https://googleapis.github.io/python-genai/) to general availability (GA).\n\n## January 21, 2025\n\n**Model updates:**\n\n- Released `gemini-2.0-flash-thinking-exp-01-21` , the latest preview version of\nthe model behind the[Gemini 2.0 Flash Thinking Model](https://ai.google.dev/gemini-api/docs/thinking) .\n\n## December 19, 2024\n\n**Model updates:**\n\n- Released Gemini 2.0 Flash Thinking Mode for public preview. Thinking Mode is a test-time compute model that lets you see the model's thought process while it generates a response, and produces responses with stronger reasoning capabilities. Read more about Gemini 2.0 Flash Thinking Mode in our [overview\npage](https://ai.google.dev/gemini-api/docs/thinking-mode) .\n\n## December 11, 2024\n\n**Model updates:**\n\n- Released [Gemini 2.0 Flash Experimental](https://ai.google.dev/gemini-api/docs/models/gemini#gemini-2.0-flash) for public preview. Gemini 2.0 Flash Experimental's partial list of features includes:\n  - Twice as fast as Gemini 1.5 Pro\n  - Bidirectional streaming with our Live API\n  - Multimodal response generation in the form of text, images, and speech\n  - Built-in tool use with multi-turn reasoning to use features like code execution, Search, function calling, and more\n\nRead more about Gemini 2.0 Flash in our [overview\npage](https://ai.google.dev/gemini-api/docs/models/gemini-v2).\n\n## November 21, 2024\n\n**Model updates:**\n\n- Released `gemini-exp-1121` , an even more powerful experimental Gemini API model.\n\n**Model updates:**\n\n- Updated the `gemini-1.5-flash-latest` and`gemini-1.5-flash` model aliases\nto use`gemini-1.5-flash-002` .\n  - Change to `top_k` parameter: The`gemini-1.5-flash-002` model supports`top_k` values between 1 and 41 (exclusive).\nValues greater than 40 will be changed to 40.\n- Change to \n\n## November 14, 2024\n\n**Model updates:**\n\n- Released `gemini-exp-1114` , a powerful experimental Gemini API model.\n\n## November 8, 2024\n\n**API updates:**\n\n- Added [support for Gemini](https://ai.google.dev/gemini-api/docs/openai) in the OpenAI libraries / REST API.\n\n## October 31, 2024\n\n**API updates:**\n\n## October 3, 2024\n\n**Model updates:**\n\n- Released `gemini-1.5-flash-8b-001` , a stable version of our smallest Gemini\nAPI model.\n\n## September 24, 2024\n\n**Model updates:**\n\n- Released `gemini-1.5-pro-002` and`gemini-1.5-flash-002` , two new stable\nversions of Gemini 1.5 Pro and 1.5 Flash, for general availability.\n- Updated the `gemini-1.5-pro-latest` model code to use`gemini-1.5-pro-002` and the`gemini-1.5-flash-latest` model code to use`gemini-1.5-flash-002` .\n- Released `gemini-1.5-flash-8b-exp-0924` to replace`gemini-1.5-flash-8b-exp-0827` .\n- Released the [civic integrity safety filter](https://ai.google.dev/gemini-api/docs/safety-settings#safety-filters) for the Gemini API and AI Studio.\n- Released support for two new parameters for Gemini 1.5 Pro and 1.5 Flash in\nPython and NodeJS:\n[`frequencyPenalty`](https://ai.google.dev/api/generate-content#FIELDS.frequency_penalty) and[` presencePenalty`](https://ai.google.dev/api/generate-content#FIELDS.presence_penalty) .\n\n## September 19, 2024\n\n**AI Studio updates:**\n\n- Added thumb-up and thumb-down buttons to model responses, to enable users to provide feedback on the quality of a response.\n\n**API updates:**\n\n- Added support for Google Cloud credits, which can now be used towards Gemini API usage.\n\n## September 17, 2024\n\n**AI Studio updates:**\n\n- Added an **Open in Colab** button that exports a prompt – and the\ncode to run it – to a Colab notebook. The feature doesn't yet support\nprompting with tools (JSON mode, function calling, or code execution).\n\n## September 13, 2024\n\n**AI Studio updates:**\n\n- Added support for compare mode, which lets you compare responses across models and prompts to find the best fit for your use case.\n\n## August 30, 2024\n\n**Model updates:**\n\n- Gemini 1.5 Flash supports\n[supplying JSON schema through model configuration](https://ai.google.dev/gemini-api/docs/json-mode#supply-schema-in-config) .\n\n## August 27, 2024\n\n**Model updates:**\n\n- Released the following\n[experimental models](https://ai.google.dev/gemini-api/docs/models/experimental-models) :\n  - `gemini-1.5-pro-exp-0827`\n  - `gemini-1.5-flash-exp-0827`\n  - `gemini-1.5-flash-8b-exp-0827`\n\n## August 9, 2024\n\n**API updates:**\n\n- Added support for [PDF processing](https://ai.google.dev/gemini-api/docs/document-processing) .\n\n## August 5, 2024\n\n**Model updates:**\n\n- Fine-tuning support released for Gemini 1.5 Flash.\n\n## August 1, 2024\n\n**Model updates:**\n\n- Released `gemini-1.5-pro-exp-0801` , a new experimental version of[Gemini 1.5 Pro](https://ai.google.dev/gemini-api/docs/models/gemini#gemini-1.5-pro) .\n\n## July 12, 2024\n\n**Model updates:**\n\n- Support for Gemini 1.0 Pro Vision removed from Google AI services and tools.\n\n## June 27, 2024\n\n**Model updates:**\n\n- General availability release for Gemini 1.5 Pro's 2M context window.\n\n**API updates:**\n\n- Added support for [code execution](https://ai.google.dev/gemini-api/docs/code-execution) .\n\n## June 18, 2024\n\n**API updates:**\n\n- Added support for [context caching](https://ai.google.dev/gemini-api/docs/caching) .\n\n## June 12, 2024\n\n**Model updates:**\n\n- Gemini 1.0 Pro Vision deprecated.\n\n## May 23, 2024\n\n**Model updates:**\n\n- [Gemini 1.5 Pro](https://ai.google.dev/gemini-api/docs/models/gemini#gemini-1.5-pro) (`gemini-1.5-pro-001` ) is generally available (GA).\n- [Gemini 1.5 Flash](https://ai.google.dev/gemini-api/docs/models/gemini#gemini-1.5-flash) (`gemini-1.5-flash-001` ) is generally available (GA).\n\n## May 14, 2024\n\n**API updates:**\n\n- Introduced a 2M context window for Gemini 1.5 Pro (waitlist).\n- Introduced pay-as-you-go [billing](https://ai.google.dev/gemini-api/docs/billing) for Gemini 1.0\nPro, with Gemini 1.5 Pro and Gemini 1.5 Flash billing coming soon.\n- Introduced increased rate limits for the upcoming paid tier of Gemini 1.5 Pro.\n- Added built-in video support to the [File API](https://ai.google.dev/api/rest/v1beta/files) .\n- Added plain text support to the [File API](https://ai.google.dev/api/rest/v1beta/files) .\n- Added support for parallel function calling, which returns more than one call at a time.\n\n## May 10, 2024\n\n**Model updates:**\n\n- Released [Gemini 1.5 Flash](https://ai.google.dev/gemini-api/docs/models/gemini#gemini-1.5-flash) (`gemini-1.5-flash-latest` ) in preview.\n\n## April 9, 2024\n\n**Model updates:**\n\n- Released [Gemini 1.5 Pro](https://ai.google.dev/gemini-api/docs/models/gemini#gemini-1.5-pro) (`gemini-1.5-pro-latest` ) in preview.\n- Released a new text embedding model, `text-embeddings-004` , which supports[elastic embedding](https://ai.google.dev/gemini-api/docs/embeddings#elastic-embedding) sizes under 768.\n\n**API updates:**\n\n- Released the [File API](https://ai.google.dev/api/rest/v1beta/files) for temporarily storing\nmedia files for use in prompting.\n- Added support for prompting with text, image, and audio data, also\nknown as *multimodal* prompting. To learn more, see[Prompting with media](https://ai.google.dev/gemini-api/docs/prompting_with_media) .\n- Released [System instructions](https://ai.google.dev/gemini-api/docs/system-instructions) in\nbeta.\n- Added\n[Function calling mode](https://ai.google.dev/gemini-api/docs/function-calling#function_calling_mode) ,\nwhich defines the execution behavior for function calling.\n- Added support for the `response_mime_type` configuration option, which lets\nyou request responses in[JSON format](https://ai.google.dev/gemini-api/docs/api-overview#json) .\n\n## March 19, 2024\n\n**Model updates:**\n\n- Added support for\n[tuning Gemini 1.0 Pro](https://developers.googleblog.com/en/tune-gemini-pro-in-google-ai-studio-or-with-the-gemini-api/) in Google AI Studio or with the Gemini API.\n\n## December 13 2023\n\n**Model updates:**\n\n- gemini-pro: New text model for a wide variety of tasks. Balances capability and efficiency.\n- gemini-pro-vision: New multimodal model for a wide variety of tasks. Balances capability and efficiency.\n- embedding-001: New embeddings model.\n- aqa: A new specially tuned model that is trained to answer questions using text passages for grounding generated answers.\n\nSee [Gemini models](https://ai.google.dev/gemini-api/docs/models/gemini) for more details.\n\n**API version updates:**\n\n- v1: The stable API channel.\n- v1beta: Beta channel. This channel has features that may be under development.\n\nSee [the API versions topic](https://ai.google.dev/gemini-api/docs/api-versions) for more details.\n\n**API updates:**\n\n- `GenerateContent` is a single unified endpoint for chat and text.\n- Streaming available through the `StreamGenerateContent` method.\n- Multimodal capability: Image is a new supported modality\n- New beta features:\n  - [Function Calling](https://ai.google.dev/gemini-api/docs/function-calling)\n  - Attributed Question Answering (AQA)\n- Updated candidate count: Gemini models only return 1 candidate.\n- Different Safety Settings and SafetyRating categories. See\n[safety settings](https://ai.google.dev/gemini-api/docs/safety-settings) for more details.\n- Tuning models is not yet supported for Gemini models (Work in progress).", "url": "https://wpnews.pro/news/gemini-2-5-models-extended-indefinitely-for-aactive-users", "canonical_source": "https://ai.google.dev/gemini-api/docs/changelog", "published_at": "2026-09-22 10:44:03+00:00", "updated_at": "2026-09-22 10:54:20.245082+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-agents", "ai-products", "generative-ai"], "entities": ["Google", "Gemini 2.5", "Gemini 3.8 Flash", "Gemini 3.5 Flash-Lite", "Gemini 3.8 Live", "Gemini 3.8 Live Extended Thinking", "Lyria 3.5", "antigravity-preview-09-2026"], "alternates": {"html": "https://wpnews.pro/news/gemini-2-5-models-extended-indefinitely-for-aactive-users", "markdown": "https://wpnews.pro/news/gemini-2-5-models-extended-indefinitely-for-aactive-users.md", "text": "https://wpnews.pro/news/gemini-2-5-models-extended-indefinitely-for-aactive-users.txt", "jsonld": "https://wpnews.pro/news/gemini-2-5-models-extended-indefinitely-for-aactive-users.jsonld"}}