Changelog

Text models3

• Venice Large

  Increased context size on Venice Large from 32k to 128k tokens.

  All users

• Venice Large

  Increased context size on Venice Large from 32k to 128k tokens.

  All users

• Simple Mode Router

  Upgraded the model router in Simple Mode to a latest variant with superior routing and added support for additional languages.

  All users

Image models4

• Default Image Model

  Changed the default image model from Venice SD3.5 to HiDream.

  All users

• Fluently

  Retired Fluently.

  All users

• FLUX Custom

  Updated FLUX Custom to utilize an improved fine-tune model.

  All users

• Stable Diffusion 3.5

  Retired Stable Diffusion 3.5.

  All users

Chat15

• Conversation Name Tooltip

  Added a tooltip to show the full conversation name in the conversations list in the left drawer.

• Hide Model Thinking Output

  Added a new Character setting to hide model reasoning or thinking output for relevant models during character conversations.

• Info Buttons

  Added (i) info buttons to text chats to view message details such as temperature and top-p.

• Inline Citations

  Added in-line citations to search results.

• Latency Timer

  Added a latency timer to the message generation in the message header.

• Reasoning Mode Toggle

  Added a toggle to enable or disable reasoning mode on supported models.

• Show 9 more

Image, video and audio7

• Image Rendering

  Images now render in the UI as they complete, instead of waiting for all four to finish.

• Progressive Image Display

  Image generation now shows each image as it’s generated instead of waiting for all to complete, resulting in faster time to first image.

• Upscale Display

  Added UI elements to display upscale/enhance images in chat history. Upscales greater than 1x are automatically downloaded.

• Upscaler Timer

  Added a timer to the upscaler message UI to indicate how long the process has been running.

• Anonymous Image Limits

  Increased anonymous user image generation limits to 16 per day.

• Error Messaging

  Clarified wording on error messages related to capacity for upscale and enhanced images.

• Show 1 more

Account and settings2

• Personal Info Masking

  Added a setting in 'App Settings' that allows users to mask their personal information in the user interface.

• Sign-up Link

  Added a sign-up link in the left navigation menu when signed out.

General5

• HEIC Image Support

  Added support for HEIC images for vision models and upscale/enhance features.

• Hover Action on Images

  Added a hover action state on images in the app.

• Image Upscaler Performance

  Improved performance of the image upscaler / enhancer.

• Rate Limit Display

  Updated the rate limit display to populate on first page load.

• Code Section

  Deprecated the code-specific section of the web app to streamline the web UI.

• Top Logprobs Support

  Added support for the top_logprobs parameter in the chat completions API.

• Multiple Image Processing

  Added support for multiple image processing to Qwen VL via API.

• Upscale API Errors

  Updated the upscale/enhance API to provide clearer error messages when images are outside supported size bounds.

• Embeddings Endpoint

  Released the embeddings endpoint for all users.

• Venice Name in Model List

  Added the Venice 'Name' to the Model List endpoint. API docs have been updated.

• API Keys Endpoint

  Optimized performance of the API keys list endpoint, especially beneficial for users with high API traffic.

• Show 1 more

• Default Settings Reset

  Fixed an issue where default settings did not reset across all models as expected.

  Chat

• Double Scroll Bars

  Fixed a bug where double scroll bars could show up on chat settings.

  Chat

• Encrypted Link Sharing Bug

  Fixed a bug that was showing a success message on shared encrypted links when the sharing was unsuccessful.

  Chat

• Escaped Dollar Signs

  Fixed a rendering bug related to escaped dollar signs in code blocks within chat.

  Chat

• Hallucinated References

  Fixed a bug that could cause Venice Small and Venice Reasoning to hallucinate references when no web search was conducted.

  Chat

• Horizontal Scrolling for LaTeX

  Fixed a bug that prevented horizontal scrolling when viewing complex LaTeX formulas wider than the viewport.

  Chat
• Show 8 more