Rivya Chat Studio: Text Conversations and Saved Chats
Send a text request in Rivya Chat Studio, check credits and saved conversations, and review prompts before moving them to image or video generation.
Last reviewed on November 11, 2026
Chat Studio is Rivya's signed-in workspace for text conversations, writing and planning. You can continue saved conversations and use replies to prepare image or video tasks. Chat does not automatically create a finished media project. For ways to organize longer work, see chat workflows.
This page covers the browser workspace. For API keys and programmatic chat requests, see the Chat API. The Public API cannot read conversations created only in Studio.
Text-only Gemini 4 Carbon requests
Gemini 4 Carbon accepts text without attachments, web search, tools or exposed reasoning settings in this Rivya path. It delivers one complete answer after terminal validation. The reserve covers the model’s full published context for both input and output; it does not use a default answer length as a billing cap.
The current promotional quote expires at 00:00 UTC on November 30, 2026. New requests stop then until a new quote is confirmed. Existing requests keep their creation-time quote and settle from confirmed consumption. Do not automatically continue or resend an uncertain request.
Send your first message
Sign in and open Chat from Studio. Select an available model and check its controls. A model listed in the directory may be paused or Coming Soon; see model availability if the one you expected is missing.
Write the task, relevant facts, and desired response format. For example: “Write three product-photo concepts for a blue ceramic mug. Keep the mug blue, use no invented product claims, and describe the setting and lighting for each.” This is a sample request, not a tested result.
Review the displayed credit estimate and the model's controls, then send once. Include a text request; the currently available chat models do not accept image attachments.
Wait for the reply, read it, and follow up in the same conversation when you want the saved exchange included. Each message can contain up to 8,000 characters. Split longer material into relevant parts, and keep your own copy of the full source.
A reply can help you draft or reason, but it does not verify the product facts you supplied. Review constraints and claims before using the answer. Prompt writing basics explains how to make a request specific.
Attach images and use image shortcuts
Image attachments and image analysis are currently unavailable in Chat Studio. The available models accept text, even if their suppliers offer other inputs elsewhere. PDFs, videos and audio recordings are not chat attachments here. See references and uploads for workspaces that accept files.
An image in an older conversation or a link into Chat does not make the selected model accept images. Use a workspace that supports your file, and check its format, size and file-count limits before uploading. Removing a pending thumbnail need not delete the stored file. Prepare a suitable copy using the safe upload guide.
Do not rely on an image shortcut to analyze a picture with the current text-only models. If an older conversation shows one of these shortcuts, clicking it sends a chat request immediately and can use credits. Tool starter suggestions, when shown, fill the text box for you to edit before sending. See tools for those task presets.
Choose model controls and understand costs
The currently available chat models offer a Thoughts setting where supported; they do not offer web search or a reasoning-effort selector. The setting does not guarantee a correct answer or a complete account of the model's reasoning. Verify facts and any cited page yourself, including its date; a generated link is not evidence that the answer is current.
The credit estimate beside the composer starts at the model's base amount. It is not a fixed price for every message. Rivya reserves credits using the selected model, saved context and expected response size. Unused reserved credits can be returned when Rivya confirms a charge within that reservation. A saved answer does not by itself confirm final credits. Longer conversations can cost more, and each follow-up or shortcut is a new request. See credits and billing.
If a turn fails, read the error and check the saved conversation and credit records before retrying. Failed tasks and refunds explains why a recorded failure differs from an answer you do not like. A lost connection or interrupted reply does not by itself confirm failure or a refund.
Continue a saved conversation or start fresh
Saved exchanges belong to your account and can be reopened from History. A conversation keeps its model and saved controls. Text that appeared during streaming, or text still in the composer, does not prove that an exchange was saved; see saved sessions. Wait for a saved conversation to load before sending a follow-up.
New chat clears the current conversation view, unsent text, and pending attachments. It does not delete earlier saved sessions. Save important drafts separately before using it, refreshing, or changing pages.
Changing models can open a different model page or clear the current conversation. It does not move a saved conversation and all its context to the new model. Copy the facts and decisions you need into a new request and check the new model's controls. Some switches carry the draft text, but keep a copy of unsent work before leaving the page.
Take a reviewed prompt into image or video generation
Where available, Use for image and Use for video open a generation form with selected text from the reply. An older conversation's nearby image reference may also be carried over; check whether the destination accepts it. The links do not transfer the full chat history or all its images, and they do not submit a generation task.
If a reply uses IMAGE_PROMPT:, NEGATIVE_PROMPT: and VIDEO_PROMPT: sections, the interface may show a prompt set with copy controls. Review the section you need. Negative instructions do not automatically fill a destination's negative-prompt field. Longer text can be carried in the link's fragment, but the destination model still has its own input limit. Compare the complete prompt you approved with the destination form before submitting.
Before generating, confirm the destination model, input mode, reference image, prompt, parameters and credit estimate. Check whether that model is suitable for your task. Follow Image Studio or Video Studio for submission. For audio, copy a script or sound description into an available model using the audio workflow. Suno Music supports new custom-mode generation. Music extension, WAV conversion and vocal separation from Music results remain unavailable.
Recover from an unclear or interrupted reply
Keep a copy of the request before retrying. If View conversation is shown, open it to check the saved exchange; you can also find the conversation in History. An interrupted stream may leave text on the page without confirming it was saved. Closing the page, a timeout or a lost connection does not prove that the request was cancelled or credits were returned. If the answer is already saved, continue from it. Sending the same request again is a new turn and can use more credits.
For page or session problems, check browser and file requirements. If the issue persists, follow troubleshooting with the model, approximate time and visible error. Share only the details needed to investigate, without login codes, API keys or confidential source material. Review factual, numerical and commercial claims before publishing a draft from Chat.
