omnivoice-rs Releases

Introduces a comprehensive rework of the `doctor` command, transforming it into a four-state audit tool with post-mortem analysis and live probing capabilities.

Jul 15, 2026

Rebuilds the `--discover` command into a structured pipeline with a new nominate, enrich, and floor workflow, and introduces global trending.

Jul 13, 2026

This release introduces significant performance improvements through support for int8 conversion models, including faster processing and Turing GPU support. It also adds new features like a Seed node and Ideogram JSON prompt support, alongside updates for various partner models.

Jun 30, 2026

This release focuses on expanding multimodal capabilities with support for Qwen3-VL and SCAIL-2 multireference, along with new partner models like Kling V3-Turbo and Luma Rays 3.2. It also includes updates to the API and UI naming conventions.

Jun 23, 2026

A minor release adding support for the Kling V3-Turbo model.

Jun 18, 2026

This release introduces significant updates to 3D visualization capabilities, assets API pagination, and partner node integrations. It includes performance optimizations for video generation models like SCAIL-2 and Wand, alongside fixes for image grids and audio node consolidation. The update also adds support for new models such as Bernini-R and Runway Aleph2.

Jun 16, 2026

This release introduces support for the Ideogram V4 model and enhances PiD capabilities with SDXL and QwenImage support. It also includes critical stability fixes for MultiGPU operations, math expression handling, and model inference for Birefnet and Dino3.

Jun 3, 2026

This release enhances Difftastic's syntax-aware diffing capabilities by correctly handling trailing punctuation and introduces support for Assembly. It also improves parsing for multiple languages and updates build requirements.

Apr 30, 2026

A draft release offering binary packages for macOS, Linux, and Windows, including support for CUDA 12.6.

Apr 22, 2026

A draft release containing binary distributions for macOS, Linux, and Windows, featuring CUDA 12.6 compatibility.

Apr 11, 2026

A draft release providing binaries for macOS (Apple Silicon and Intel), Linux, and Windows, utilizing CUDA 12.6.

Apr 11, 2026

A CI infrastructure release containing necessary assets (reference artifacts, live oracle assets, and audio files) for the macOS Metal workflow.

Apr 10, 2026

This release introduces support for vision modality in LLM interactions and migrates the LLM API caching mechanism from pickle to JSON.

Mar 28, 2026

This release focuses on improving Git integration, updating parsers for various programming languages, and refining build requirements and display logic.

Mar 16, 2026

This release updates the default model to GPT-5-mini and introduces tools for validating simulations against empirical data, alongside new cost tracking and Jupyter notebook integration.

Feb 2, 2026

This release focuses on enhancing LLM integration with automatic model loading and connection testing, while fixing specific bugs related to digital avatar parameter passing and workflow setup.

Jan 27, 2026

Introduces new content creation modules including action migration and expands local workflow support for digital avatars and image-to-video generation.

Jan 26, 2026

Launches core video generation pipelines for digital avatars and audio-visual content, supported by multi-language TTS support and expanded hardware compatibility for 48G GPUs.

Jan 14, 2026

Enhances configuration flexibility with API key support and concurrency controls, while improving LLM stability, system configuration hot-reloading, and UI template updates.

Dec 28, 2025

Improves user experience with built-in help documentation and adds new templates, specifically addressing TTS stability issues caused by edge-tts version compatibility.

Dec 10, 2025