agency-swarm Releases
139 releases of VRSEN/agency-swarm
- v1.9.5
Patch release addressing FastAPI LiteLLM client configuration handling, specifically warning and dropping invalid keys when LiteLLM is missing.
Apr 29, 2026
- v1.0.71
EXO v1.0.71 is a patch release that adds the Kimi K2.6 model, improves sampling defaults, and fixes bugs related to M5 Macs and RDMA.
Apr 23, 2026
- v1.9.4
Patch release focusing on Agent Swarm CLI compatibility and fixing Codex API event handling.
Apr 22, 2026
- v1.9.3
Patch release for provider-scoped FastAPI client configuration, ensuring `client_config.base_url` is scoped correctly by LiteLLM provider.
Apr 21, 2026
- v1.9.2
Patch release for Agent Swarm CLI TUI output containment and introducing per-request agent model overrides via FastAPI.
Apr 18, 2026
- v1.0.70
EXO v1.0.70 is a major release introducing multimodality support and new model support, alongside performance improvements like Flash Attention.
Apr 17, 2026
- v1.9.1
Patch release addressing browser-auth errors, CLI TUI output containment, and CLI session-memory compatibility.
Apr 17, 2026
- 0.2.6
This release updates the faster-qwen3-tts library to allow `None` values for the `non_streaming_mode` parameter, preserving upstream defaults.
Apr 17, 2026
- v1.9.0v1.9.0 - Agent Swarm TUI and OpenClaw
Major release introducing Agent Swarm TUI, OpenClaw support, and request-scoped FastAPI control.
Apr 16, 2026
- 2.0.2v2.0.2: fixed some bugs
This is a minor release focused on fixing bugs identified in the previous version.
Apr 8, 2026
- 2.0.1v2.0.1: removed auto-trim feature for reference audio
A hotfix release that removes the auto-trim feature for reference audio.
Apr 8, 2026
- 2.0.0VoxCPM 2 Released
VoxCPM 2 is released with a collection of bug fixes, performance optimizations, and dependency updates.
Apr 6, 2026
- 0.2.5
A comprehensive update to faster-qwen3-tts including fixes for CUDA graph reliability, a new instruct parameter for voice cloning, and improvements to streaming playback and documentation.
Mar 27, 2026
- v1.0.69
A significant update adding Qwen3.5 support, continuous batching, M5 Pro/Max chip support, and a host of UI improvements and bug fixes.
Mar 27, 2026
- v0.8.3
This release includes dependency updates, the addition of OpenTelemetry observability, an interactive agent loop mode, and fixes for the web search tool loading when API keys are in config files.
Mar 23, 2026
- v1.0.68
A major update focusing on extensive model support (HuggingFace, Claude, OpenAI), API enhancements, and critical stability fixes for GPU locks.
Feb 25, 2026
- v1.8.0v1.8.0 — Present and Accounted For
This release improves first-message UX, cross-provider history compatibility, and metadata visibility while tightening LiteLLM release validation and upgrading the OpenAI Agents SDK. It includes better defaults, cleaner migration paths, and safer release behavior.
Feb 25, 2026
- v0.8.2
A minor update focusing on dependency bumps, documentation fixes, and a new feature for human-in-the-loop proxy port exposure.
Feb 24, 2026
- v0.8.1
Resolves LLM API base and model configuration issues while normalizing alternative tool call formats.
Feb 20, 2026
- v0.8.0
A major update including LiteLLM upgrades for new provider support, a redesigned vulnerability reporting system, and TUI improvements.
Feb 19, 2026