jitsu Releases

May 15, 2026

This release introduces 'Bring Your Own LLM Infrastructure' with a new Custom LLM provider, ambient audio for live calls, and Assistant Authentication to verify sessions before agents start.

May 13, 2026

This release focuses on the setup and quick start guide for Electrobun, detailing how to install it via the `bun` package manager and how the CLI handles automatic artifact downloading.

May 4, 2026

This release fixes a regression in the device-monitor component that was broken in the previous version.

Apr 28, 2026

This beta release focuses on making Inbound Voice AI production-ready by introducing distributed SIP registration, multi-server failover, and enhanced transfer tooling.

Apr 28, 2026

This release addresses a critical security vulnerability and introduces a new environment variable to the configuration files.

Apr 27, 2026

This release resolves a launch issue with vLLM versions greater than 0.18 when running on systems with multiple GPUs.

Apr 17, 2026

This release addresses a critical security vulnerability and strongly recommends immediate upgrading for all users to ensure system security.

Apr 12, 2026

A major release introducing built-in observability and a streaming pipeline architecture rewrite to improve performance, monitoring, and developer experience.

Apr 1, 2026

This is a maintenance release focusing on bug fixes for reporting and transaction handling, alongside UX improvements for forms and codebase formatting.

Mar 28, 2026

Focused on smarter listening and reliability, this release introduces pluggable VAD/EOS engines and comprehensive test coverage across multiple providers.

Mar 17, 2026

A feature-rich SDK release adding multiple export formats for reports, expanding integrations (Payment Links, Stripe, Plaid), and refactoring the ExchangeRates module to NestJS.

Mar 16, 2026

This release introduces support for HAMi DRA and leader election among schedulers, alongside optimizations for NVIDIA device plugins and the HAMi webUI. It also addresses numerous bug fixes including vLLM issues, quota calculation errors, and concurrent map errors, while improving support for iluvatar and Ascend GPUs.

Jan 20, 2026

Open-Sora V1.3 introduces improvements in model architecture and generation quality, while maintaining support for image-to-video and video extension with expanded resolution and frame length.

Feb 21, 2025

The v0.0.55 release removes the TensorRT dependency from the megaparse library.

Feb 14, 2025

The megaparse SDK v0.1.12 adds layout detection functionality to the library.

Feb 14, 2025

Version 0.1.11 introduces layout detection and text detection capabilities within the auto strategy, alongside bug fixes for EngineConfig and StrategyHandler.

Feb 11, 2025

The v0.0.54 release adds layout detection functionality to the core megaparse library.

Feb 11, 2025

Version 0.0.53 introduces a modular parser and formatter system, adds text detection to the auto strategy, and implements type strategy output, accompanied by various bug fixes.

Jan 16, 2025

Open-Sora v1.2 is released, building upon v1.1 with new components to enhance video generation capabilities, including compression networks and improved training flows.

Jun 18, 2024