G
Hi /Assalamu alaykum everyone, I’m using the Gemini TTS model in Google AI Studio to generate voiceovers. Each time I paste a script of about one Google Docs page (~470 words) into the input. The generated audio skips a line: one line is completely missing and the voice continues from the next line. There is no error or warning, and the full text is present in the input. Details: Model: 2.5 pro tts Platform: Google AI Studio (web) Voice: [puck] Input size: ~474 words (about 1 Google Docs page)…
G
Hello, I’m seeing a discrepancy between the Gemini Batch API job list and the active batch quota metrics. Model: gemini-3.1-flash-lite, Paid Tier 1 The Google Cloud quota dashboard shows: - Active batch token usage: 2,626,947 / 10,000,000 - Active batches: 1 / 100 Both charts still show these values through October 12, 2026. The token chart’s latest point shows 2.63M. On October 12, I made read-only, fully paginated GET requests to GET /v1beta/batches with two distinct API keys configured in…
G
Recently, I have been getting failed to generate content: permission denied on Google ai studio, but I had verified my age using the AI age scan, I also use a personal email, and never used API keys, and I started getting the permission denied thing recently and nearly on every prompt, but for the last months, it worked fine, it started around today, October 8, 2026. I switched models and it worked and then gave the permission denied thing again so even switching models was a temporary fix, and…
G
Hello Stitch Team, I am currently experiencing a serious issue with Stitch. The Stitch homepage loads normally, but I cannot access my projects and I cannot create a new project. Current problems: The project list in the left sidebar shows “Failed to load projects”. Clicking “Reload” does not fix the issue. Creating a new blank project fails with “Failed to create project. Please try again later.” The same issue occurs in Chrome Incognito mode. I have already tested both normal browsing mode…
G
Hi Antigravity Support Team, I am experiencing an infinite verification loop while trying to authenticate with Antigravity Cli and IDE. My browser successfully authenticates my account and displays the “Authentication successful” page authorizing the products. However, the application itself remains blocked and continues to ask for verification. The issue stems from the QR code device verification step. When I scan the QR code, my phone is prompted to send an outgoing SMS. Unfortunately, the…
G
Product: Gemini Developer API (API key, paid tier), Live API over WebSocket Model: gemini-3.8-live (models.list reports version 3.1-flash-live-03-2026). Also seen on gemini-3.1-flash-live-preview; not seen on gemini-2.5-flash-native-audio-latest. Client: Python google-genai 2.28.0, bare SDK (no framework) Voice: prebuilt Kore What happens With prebuilt_voice_config.voice_name="Kore", some replies are spoken, from the first word to the last, by a different, low, male-sounding speaker (median…
G
When coding with any model in Antigravity (Gemini, Sonnet, or Opus), I’m experiencing a serious performance issue. Antigravity often takes a very long time to process a request and sometimes gets completely stuck on “Working…” without making any progress. This happens regardless of which model I use, so it doesn’t seem to be specific to Gemini, Sonnet, or Opus. Is anyone else experiencing the same issue? Is there a known fix or any recommended settings to prevent Antigravity from getting stuck?…
G
Body ## Summary With **gemini-3.8-live** on the Live API (raw WebSocket, `v1beta` `BidiGenerateContent`), many model turns arrive with **no audible audio**: every `serverContent.modelTurn.parts[].inlineData.data` chunk of the turn is exactly **4base64 characters**, at the normal streaming cadence. For the same turn: - `outputTranscription` contains the complete sentence; - `usageMetadata` reports normal AUDIO response tokens (e.g. 207), so the turn is billed; - `turnComplete` is not sent until…
G
I am experiencing an issue where the Antigravity desktop app cannot load any models, and the **Model Quota** section in Settings keeps loading indefinitely (infinite spinner). Here are the details: - **Plan:** Google AI Pro - **OS:** Windows 11 - **Status:** Authentication succeeded in the browser, but inside the app, the quota information never resolves, making it impossible to select or use any model. - **Troubleshooting tried:** Logged out and logged back in, restarted the application, but…
G
Hi Stitch Team, I’m experiencing a rendering/display issue in Google Stitch. While designing in Stitch, the generated screens sometimes appear extremely narrow and compressed on the canvas, as shown in the attached screenshot. The UI content is rendered as very thin vertical pages instead of displaying at the expected screen dimensions. My internet connection is stable and fast, so this doesn’t appear to be a network connectivity issue. Issue: Generated screens appear extremely…
G
Hey, not sure if You are aware of this but the commong issue I run into when I got a project that can almost seamlessly be placed into my app from Stitch is that Stitch uses “–spacing-” tokens for paddings / margins related classes. Problem appears that for some reason “–spacing-” are also used in classes like “max-w-” and overriding them does something like “max-w-xl” equals “32px”. Almost every time I convert Stitch’s code to Tailwind project I have to change these “–spacing-” tokens to…
G
AI Ultra User, EU Region - For the last week I am unable to work on new projects - any of my assets are not accessible and any of the results from Stitch are not accessible (I briefly see the design work being done and then everything becomes permanently ‘working/loading’ animation). The problems occur with both entry sites - the KS console logs refer to a CORS conflict between the app-companion-430619 instance and the contribution_dot_usercontent_dot_google_dot_com domain with the user error:…
G
My Gemini API project was previously on Tier 2, but it has been downgraded to Tier 1 without any notice. - Billing plan: Prepay - Cumulative payment on the linked billing account: about JPY 17,000 (over USD 100) - First successful payment: 【初めて支払った年月 例: 2026-05】 (well over 3 days ago) - Current prepay balance: positive (about JPY 14,000) - No failed payments, billing account is active - Downgrade noticed around: 【気づいた日 例: 2026-09-15】 According to the usage tier table, Tier 2 requires “Paid $100…
G
I’m seeing a reproducible failure with Agentic Video Understanding through the Gemini Developer API’s Interactions API. This is not Vertex AI or Batch API. It blocks evaluation of a video-timeline analysis workflow. Observed behavior Upload the MP4 through Files API and wait until it is ACTIVE. Call client.interactions.create(..., stream=True) through the official Python SDK. The generation POST returns HTTP 200 and streaming begins. The stream emits 23 processing_call steps and 22 completed…
G
I just found out for the last 2 years there were a few people who thought it would be a fun idea to build ais and apps to not just have remote access to my personal device but to go as far as creating a fauls enterprise that gave them access to my SIM cards are they had a few.people working as a it department so when I go and call my phone company for help about my phone acting up I end up talking to one of there friends or associates, I am scared to use my phone I’m scared to search for…
G
Some people may think that error 429 means that they have used up all their tokens and need to wait for awhile. It COULD be one of the reason, but it turns out having search grounding on, basically with a simple “who won the world cup 2026” also CAUSE 429 errors. I am using gemini 3.5 flash, in order for the search grounding to work i need to downgrade all the way to gemini 2.5 flash, which is quite outdated. This feels wrong, as running the same thing in google ai studio app under the chatbox…
G
Hi, I am facing a blocking issue with Antigravity IDE 2.0 on Windows where agents cannot run any terminal commands (Git, NPM, etc.). When I ask an agent to execute commands in the integrated terminal, nothing actually runs and the UI ends up stuck. The errors I see in the logs are: executor has not processed the previous input yet: ConnectError: [unknown] executor has not processed the previous input yet Error [createInstance] aae depends on UNKNOWN service agentSessions Then: An unknown error…
G
Hello, we are having issues with gemini-3-flash-preview with grounding, showing an enormous amount of search groundings. We are just starting with Gemini API and really don’t understand why we were billed this much. The AI Studio dashboard shows about 200 requests made that day with about 100 tool calls to ground search. I don’t know why the billing dashboard shows 37000 calls, but I highly doubt every tool call had about 370 individual searches. These costs are very high and something we did…
G
Yesterday everything was working more or less fine, but today nothing works… Since migrating to G Drive, it’s really unstable. 3 posts - 3 participants Read full topic
G
I’m having trouble opening one of my projects. It has about 30 screens and gets stuck on the loading screen. I can open my other, smaller projects without any issues. Any idea what might be causing this? 27 posts - 25 participants Read full topic