docs(changelog): add Coder Daemon Deepseek, Fast Context, and Kimi fixes

This commit is contained in:
2026-05-26 20:09:32 +02:00
parent 635cac22fe
commit 126ffa1fb0
+3
View File
@@ -68,6 +68,8 @@ import BaseLayout from '../layouts/BaseLayout.astro';
</h3>
<ul class="changes-list">
<li><strong>Zaguán Coder Daemon.</strong> Added full support for versioned Full Context.</li>
<li><strong>Zaguán Coder Daemon.</strong> Updated the system prompt and logic for Deepseek V4 Pro/Flash to make it more reliable and to take action quicker.</li>
<li><strong>Zaguán Coder Daemon.</strong> Properly wired up the new Fast Context logic and added instructions in the prompts.</li>
<li><strong>Zaguán Coder Daemon.</strong> WIP. Working on fixing the prompts and logic for each supported model.</li>
</ul>
</div>
@@ -78,6 +80,7 @@ import BaseLayout from '../layouts/BaseLayout.astro';
Bug Fixes
</h3>
<ul class="changes-list">
<li><strong>Zaguán Coder Daemon.</strong> Fixed bugs that prevented Kimi K2.6 from properly displaying its reasoning.</li>
<li><strong>Zaguán Coder Daemon.</strong> Fixed system prompts and logic so that models get to work faster instead of fetching content forever.</li>
<li><strong>Zaguán Coder Daemon.</strong> Disabled <a href="https://docs.cohere.com/docs/command-a-plus" target="_blank" rel="noopener">Cohere Command A+</a> for now due to issues that have been impossible to solve.</li>
<li><strong>Zaguán Coder Daemon.</strong> Disabled Inception Mercury 2 for now due to its smaller context window of 128k.</li>