v0.8.4 v0.2.11

Name: artemis
Author: google

medium risk

google/artemisAug 2, 2026 → Aug 3, 20261 day between releases

What's New

Adds chat-completions backend and text-only response support

Breaking Changes

Skill sources
Nextcloud Talk
Generic webhook authentication

New Features

Chat-completions LLM backend (OpenAI /v1/chat/completions)
Text-only and out-of-band response support
Realtime web demo app under demo/
Default Qwen3 TTS to GGML format

Improvements

Bump actions group dependencies
Emit assistant transcript for fresh response when discard guard is stuck
Update HF router Gemma chat completions example
Refresh README docs

Bug Fixes

Fix two mid-generation conversation races (tool call & image)
Defer client conversation items during an active response
Fix Docker image build
Remove deprecated melo option from tts arguments

Upgrade Notes

Configure the new chat-completions backend and ensure Docker image builds correctly. Review breaking changes in legacy integrations if applicable.

The Two Releases

v0.8.4

Aug 2, 2026· 3 months ago

A maintenance and hardening release expanding memory and SOP control planes, improving provider reliability, and strengthening sandbox and credential boundaries.

Full release notes

v0.2.11

Aug 3, 2026· 3 months ago

A release focused on backend improvements including support for text-only responses, a new chat-completions LLM backend, and a new realtime web demo app.

Full release notes