bytebot Releases

Speech recognition library release focusing on TTS improvements including KittenTTS support, Supertonic3 models, and various WASM and streaming fixes. Note: This release is from a different repository (k2-fsa/sherpa-onnx).

May 13, 2026

Big-AGI release introducing advanced AI models (GPT-5.5, DeepSeek V4, Anthropic Opus 4.7), improved container management, and enhanced Gemini Deep Research with resumable sessions.

May 13, 2026

Hysteria app release focused on NAT traversal improvements, specifically fixing symmetric NAT connection issues and enhancing SNI guard configuration for self-signed certificates.

May 10, 2026

Minor maintenance release with cargo clean fix and version bump to 0.11.65. This release focuses on cleaning build artifacts and preparing for the next development cycle.

May 10, 2026

Hysteria app release introducing Realms feature for NAT traversal, allowing server hosting without public IP through P2P connections.

May 10, 2026

Major TTS-focused release introducing Deepgram TTS support, enhanced connection state checks, and improved voice configuration for multiple TTS providers including Rime, Deepgram, Tencent, and Murf.

May 8, 2026

Speech recognition library release focusing on sense-voice improvements and bug fixes including bounds checks for speaker diarization and various stability improvements.

May 8, 2026

Pocket-TTS library release focusing on voice synthesis improvements including device support enhancements, default voice configuration, and voice cloning fixes.

May 4, 2026

Fixed PostgreSQL reference documentation by correcting invalid column references in query examples.

May 1, 2026

Added canonical documentation backlinks to every SKILL.md file for SEO purposes on skill aggregators.

Apr 29, 2026

Hysteria app security-focused release addressing QUIC packet handling vulnerabilities, nftables compatibility issues, and thread safety problems in obfuscation.

Apr 26, 2026

Added curated cross-references between hub skills and improved documentation workflow with Prettier formatting and explicit command names.

Apr 21, 2026

A major version bump introducing multilingual TTS models (Italian, German, Spanish, Portuguese, French) and an improved English model. It includes CLI argument updates for language selection and performance optimizations via quantization.

Apr 21, 2026

A major update integrating deeply with the Hugging Face ecosystem, supporting 20 models with Transformers as the backend. It also introduces document format conversion to Markdown and DOCX export capabilities, plus a new browser inference SDK.

Apr 21, 2026

A patch release for PaddleOCR-VL that expands hardware compatibility by adding AMD and Intel Arc GPU support. It also fixes dependency issues for Huawei NPU and KunlunXin XPU, and removes the page limit on Docker Compose services.

Apr 14, 2026

This release focuses on enhancing speech capabilities by adding Oracle Speech AI and Blaze extensions for ASR/TTS, alongside fixes for Azure and Tencent services. It also introduces metrics tracking and updates exception handling in websockets.

Apr 2, 2026

This release contains critical server stability fixes regarding congestion control and firewall operations, including a new environment variable for configuring the firewall backend.

Mar 31, 2026

This beta release addresses stability issues in the web app, specifically fixing functionality for users who haven't set a chat model and correcting the display of the deprecation banner on Khoj cloud.

Mar 26, 2026

This release updates web app page templates to ensure compatibility with Starlette 1.0.0, resolving potential rendering issues.

Mar 25, 2026

This release improves the web app experience by upgrading to Next.js 15 and adds a deprecation notice for Khoj cloud. It also includes numerous fixes for research agents, PDF processing, and memory leaks.

Mar 25, 2026