Skip to content

Releases: adamlwalker/DARIA

v2.3.1: chat model can call image gen; English-only UI

Choose a tag to compare

@adamlwalker adamlwalker released this 10 Sep 23:19
With a model loaded, Image gen is a chat tool: the model writes the
Z-Image-Turbo prompt and the PNG is attached to the reply. Without a
model, the user's message is still the prompt.

The language switcher is gone; production agent and tray stay English.

Daria 2.3.0

Choose a tag to compare

@adamlwalker adamlwalker released this 07 Sep 02:13
v2.3.0: Qwen 3.8 coding, prompt-cache reuse, and speculative decode

Bring the coding agent in line with upstream Chaty 2.1.7 without TikTok
or extra UI languages. Tool results use the tool role, reasoning stays
in its own field, and per-turn context no longer busts the KV cache —
so Qwen 3.8 long runs resume instead of re-reading the conversation.
llama.cpp moves to a Muse-Glimmer build with optional MTP speculative
decoding. Browser clicks land on the named control, generations survive
a webview reload, and the system prompt no longer embeds the clock.

Daria 2.1.0

Choose a tag to compare

@adamlwalker adamlwalker released this 16 Aug 21:35
08c13a4

First release as Daria. Discrete AI for Reasoning, Interaction & Automation. Still updating docs Daria does offer the alternative of using edge TTS for voice generation. There is a warning that that is not an off-line tool..